# RAW Works > Full-content export of the currently published Hugo site for LLM use. Generated: 2026-09-09T11:02:27Z. Base URL: https://raw.works/. Pages included: 139. By default, Hugo excludes draft, future, and expired content unless built with `-D`, `-F`, or `-E`. ## Sitemap - [Machine Studying - The Library and The Lab](https://raw.works/machine-studying-the-library-and-the-lab/): content/works/machine-studying-the-library-and-the-lab.md | published: 2026-06-20 - [Engineering Skills](https://raw.works/engineering-skills/): content/works/engineering-skills.md | published: 2026-05-12 - [Code Execution as Reasoning](https://raw.works/code-execution-as-reasoning/): content/works/code-execution-as-reasoning.md | published: 2026-05-11 - [RLMs are the new reasoning models](https://raw.works/rlms-are-the-new-reasoning-models/): content/works/rlms-are-the-new-reasoning-models.md | published: 2026-04-20 - [RLMs are SOTA on LongCoT](https://raw.works/rlms-are-sota-on-longcot/): content/works/rlms-are-sota-on-longcot.md | published: 2026-04-19 - [LongCoT - A benchmark worthy of a RLM's attention](https://raw.works/longcot-a-benchmark-worthy-of-a-rlms-attention/): content/works/rlm-longcot-sonnet.md | published: 2026-04-16 - [mycelium - an underground information network for agents](https://raw.works/mycelium-an-underground-information-network-for-agents/): content/works/mycelium.md | published: 2026-03-27 - [ypi: a recursive coding agent](https://raw.works/ypi-a-recursive-coding-agent/): content/works/ypi.md | published: 2026-02-12 - [Recursive Language Models as Memory Systems](https://raw.works/recursive-language-models-as-memory-systems/): content/works/recursive-language-models-as-memory-systems/index.md | published: 2026-02-11 - [Inversion of Caution](https://raw.works/inversion-of-caution/): content/works/inversion-of-caution.md | published: 2026-02-04 - [Introducing dirpack](https://raw.works/introducing-dirpack/): content/works/introducing-dirpack.md | published: 2026-02-02 - [Coming Soon](https://raw.works/teasers/): content/teasers/_index.md | published: 2026-01-20 - [Claude Opus 4.5 One-Shots My Crappy Handwriting](https://raw.works/claude-opus-4.5-one-shots-my-crappy-handwriting/): content/works/opus-reads-my-handwriting.md | published: 2025-12-04 - [The Zero Employee Company](https://raw.works/the-zero-employee-company/): content/works/zero-employee-company.md | published: 2025-08-19 - [No API, No Problem](https://raw.works/no-api-no-problem/): content/works/no-api-no-problem.md | published: 2025-02-20 - [Agents Love APIs](https://raw.works/agents-love-apis/): content/works/agents-love-apis.md | published: 2025-02-19 - [The Joke of the Day](https://raw.works/the-joke-of-the-day/): content/works/jokeoftheday.md | published: 2025-02-18 - [Toughened by Time](https://raw.works/toughened-by-time/): content/works/toughened-by-time.md | published: 2023-12-23 - [SDF Solstice](https://raw.works/sdf-solstice/): content/works/sdf-solstice.md | published: 2023-12-22 - [Learn Software 10x Faster With AI](https://raw.works/learn-software-10x-faster-with-ai/): content/works/learn-software-10x-faster-with-ai.md | published: 2023-12-21 - [AI-Accelerated Reading](https://raw.works/ai-accelerated-reading/): content/works/ai-accelerated-reading.md | published: 2023-12-20 - [Squeezed Into Essentialism](https://raw.works/squeezed-into-essentialism/): content/works/squeezed-into-essentialism.md | published: 2023-12-18 - [Disclosures](https://raw.works/disclosures/): content/disclosures.md | published: 2023-12-18 - [Social Media Repeaters](https://raw.works/social-media-repeaters/): content/works/social-media-repeaters.md | published: 2023-12-18 - [My COVID Protocol](https://raw.works/my-covid-protocol/): content/works/my-covid-protocol.md | published: 2023-12-17 - [LLM Prompting Goldmine](https://raw.works/llm-prompting-goldmine/): content/works/prompting-goldmine.md | published: 2023-12-16 - [Introducing: Human Instruct Turbo](https://raw.works/introducing-human-instruct-turbo/): content/works/introducing-human-instruct-turbo.md | published: 2023-12-15 - [Private and Personalized AI Transcription](https://raw.works/private-and-personalized-ai-transcription/): content/works/private-and-personalized-ai-transcription.md | published: 2023-12-14 - [What to work on? A trifecta approach](https://raw.works/what-to-work-on-a-trifecta-approach/): content/works/generative-ai-for-ecommerce.md | published: 2023-12-13 - [Zone of Genius Enumeration](https://raw.works/zone-of-genius-enumeration/): content/works/zone-of-genius-enumeration.md | published: 2023-12-12 - [Incentive Structures](https://raw.works/incentive-structures/): content/works/incentive-structures.md | published: 2023-12-11 - [The Pace Is Exhausting](https://raw.works/the-pace-is-exhausting/): content/works/the-pace-is-exhausting.md | published: 2023-12-10 - [AI Transcription Without a Subscription](https://raw.works/ai-transcription-without-a-subscription/): content/works/ai-transcription-without-a-subscription.md | published: 2023-12-09 - [Modular Frankenstein](https://raw.works/modular-frankenstein/): content/works/modular-frankenstein.md | published: 2023-12-08 - [Big Companies Love Big Data](https://raw.works/big-companies-love-big-data/): content/works/big-companies-love-big-data.md | published: 2023-12-07 - [Happy Birthday polySpectra](https://raw.works/happy-birthday-polyspectra/): content/works/happy-birthday-polyspectra/index.md | published: 2023-12-06 - [Time for Play](https://raw.works/time-for-play/): content/works/time-for-play.md | published: 2023-12-05 - [Trauma-Informed Marketing](https://raw.works/trauma-informed-marketing/): content/works/trauma-informed-marketing.md | published: 2023-12-03 - [AI as Electricity](https://raw.works/ai-as-electricity/): content/works/ai-as-electricity.md | published: 2023-12-03 - [6 Hats Is More Fun Together](https://raw.works/6-hats-is-more-fun-together/): content/works/6-hats-is-more-fun-together.md | published: 2023-12-02 - [Oh That's Why It's So Hard](https://raw.works/oh-thats-why-its-so-hard/): content/works/oh-thats-why-its-so-hard.md | published: 2023-12-01 - [Fill the Gap With Gold](https://raw.works/fill-the-gap-with-gold/): content/works/fill-the-ditch-with-gold.md | published: 2023-11-30 - [Custom Function Calling Oh My](https://raw.works/custom-function-calling-oh-my/): content/works/custom-function-calling-oh-my.md | published: 2023-11-29 - [Training Myself to Talk to AI](https://raw.works/training-myself-to-talk-to-ai/): content/works/training-myself-to-talk-to-ai.md | published: 2023-11-28 - [Fixing Typos With AI for Speed and Focus](https://raw.works/fixing-typos-with-ai-for-speed-and-focus/): content/works/fixing-typos-with-ai-for-speed-and-focus.md | published: 2023-11-27 - [Brain Retraining Question for Chronic Pain](https://raw.works/brain-retraining-question-for-chronic-pain/): content/works/brain-retraining-question-for-chronic-pain.md | published: 2023-11-26 - [Check Out Tavily](https://raw.works/check-out-tavily/): content/works/check-out-tavily.md | published: 2023-11-25 - [Family Is Where Things Don't Need to Make Sense](https://raw.works/family-is-where-things-dont-need-to-make-sense/): content/works/family-is-where-things-dont-need-to-make-sense.md | published: 2023-11-24 - [Sometimes You Get Lucky](https://raw.works/sometimes-you-get-lucky/): content/works/sometimes-you-get-lucky.md | published: 2023-11-22 - [The Big Companies Will Never Catch Up](https://raw.works/the-big-companies-will-never-catch-up/): content/works/the-big-companies-will-never-catch-up.md | published: 2023-11-22 - [Apologies for the Delay](https://raw.works/apologies-for-the-delay/): content/works/apologies-for-the-delay.md | published: 2023-11-21 - [Impromptu DJ Set at the Berkeley Half Marathon](https://raw.works/impromptu-dj-set-at-the-berkeley-half-marathon/): content/works/impromptu-dj-set-at-the-berkeley-half-marathon.md | published: 2023-11-20 - [Slow Down to Speed Up (or a Rabbit Hole)](https://raw.works/slow-down-to-speed-up-or-a-rabbit-hole/): content/works/slow-down-to-speed-up-or-a-rabbit-hole.md | published: 2023-11-19 - [Double Graduation Day](https://raw.works/double-graduation-day/): content/works/double-graduation-day.md | published: 2023-11-17 - [LLM Ensembles - A Preview](https://raw.works/llm-ensembles-a-preview/): content/works/llm-ensembles.md | published: 2023-11-17 - [Quantity: Check ✅, Quality: In Progress ⌛](https://raw.works/quantity-check-quality-in-progress/): content/works/quantity-check-quality-in-progress.md | published: 2023-11-16 - [Deep Work 1,2 Punch](https://raw.works/deep-work-12-punch/): content/works/deep-work-1-2-punch.md | published: 2023-11-15 - [The Prompting Course I Wish I Found Months Ago](https://raw.works/the-prompting-course-i-wish-i-found-months-ago/): content/works/the-prompting-course-i-wish-i-found-months-ago.md | published: 2023-11-14 - [Something for Everybody](https://raw.works/something-for-everybody/): content/works/something-for-everyone.md | published: 2023-11-13 - [Field Notes from Today's Social Media Automation Attempts](https://raw.works/field-notes-from-todays-social-media-automation-attempts/): content/works/fieldnotesfromsocialautomation.md | published: 2023-11-12 - [First Day with Athena](https://raw.works/first-day-with-athena/): content/works/first_day_with_athena.md | published: 2023-11-11 - [Cutting Edge or Bleeding Edge?](https://raw.works/cutting-edge-or-bleeding-edge/): content/works/cuttingedgeorbleedingedge.md | published: 2023-11-09 - [Digital to Physical and Back Again: Part 1](https://raw.works/digital-to-physical-and-back-again-part-1/): content/works/digitaltophysicalandbackagain.md | published: 2023-11-09 - [How Quick is Your Quickstart Guide?](https://raw.works/how-quick-is-your-quickstart-guide/): content/works/quickstart.md | published: 2023-11-08 - [Poe Wants to Become the YouTube of AI Bots](https://raw.works/poe-wants-to-become-the-youtube-of-ai-bots/): content/works/poewantstobetheyoutubeofaibots.md | published: 2023-11-07 - [Offset Monster Is Born](https://raw.works/offset-monster-is-born/): content/works/offsetmonsterisborn.md | published: 2023-11-06 - [Poe Hackathon at AGI House](https://raw.works/poe-hackathon-at-agi-house/): content/works/poehackathonatAGIhouse.md | published: 2023-11-04 - [Stuck, So Close](https://raw.works/stuck-so-close/): content/works/stucksoclose.md | published: 2023-11-03 - [Stripe Is Not for Wholesale](https://raw.works/stripe-is-not-for-wholesale/): content/works/stripe-not-for-wholesale.md | published: 2023-11-02 - [RAG Information Overload](https://raw.works/rag-information-overload/): content/works/RAG-information-overload.md | published: 2023-11-01 - [Happy Halloween 2023](https://raw.works/happy-halloween-2023/): content/works/happyhalloween2023.md | published: 2023-11-01 - [Candy-Coated Muertos](https://raw.works/candy-coated-muertos/): content/works/candycoatedmuertos.md | published: 2023-10-30 - [Enumeration of Logical Families](https://raw.works/enumeration-of-logical-families/): content/works/enumerationoflogicalfamilies.md | published: 2023-10-29 - [Social Flexural Modulus](https://raw.works/social-flexural-modulus/): content/works/socialflexuralmodulus.md | published: 2023-10-28 - [Money Can't Buy You Enrollment](https://raw.works/money-cant-buy-you-enrollment/): content/works/moneycantbuyyouenrollment.md | published: 2023-10-27 - [BREAKING NEWS: My Internet is Out](https://raw.works/breaking-news-my-internet-is-out/): content/works/breakingnewsnointernet.md | published: 2023-10-26 - [Poe Is My New Search Engine](https://raw.works/poe-is-my-new-search-engine/): content/works/poeismynewsearchengine.md | published: 2023-10-25 - [pSai and Resina](https://raw.works/psai-and-resina/): content/works/psaiandresina.md | published: 2023-10-25 - [My Journey with Cursor: A New Wave in Coding](https://raw.works/my-journey-with-cursor-a-new-wave-in-coding/): content/works/newpostaboutcursor.md | published: 2023-10-23 - [Finding the Perfect Voice Transcription Tool](https://raw.works/finding-the-perfect-voice-transcription-tool/): content/works/voice2text.md | published: 2023-10-22 - [It's All in the Hips](https://raw.works/its-all-in-the-hips/): content/works/itsallinthehips.md | published: 2023-10-21 - [GPT-4V(ision) Handwriting OCR](https://raw.works/gpt-4vision-handwriting-ocr/): content/works/gpt4v-handwritingOCR.md | published: 2023-10-20 - [Using AI at Every Step of the Customer Journey](https://raw.works/using-ai-at-every-step-of-the-customer-journey/): content/works/aiforcustomerjourney.md | published: 2023-10-19 - [Searching for Super-Discernment](https://raw.works/searching-for-super-discernment/): content/works/superdiscernment.md | published: 2023-10-18 - [Remembering Roland Griffiths](https://raw.works/remembering-roland-griffiths/): content/works/rememberingroland.md | published: 2023-10-17 - [Leveraging AI's Alexithymia to Overcome Emotional Friction at Work](https://raw.works/leveraging-ais-alexithymia-to-overcome-emotional-friction-at-work/): content/works/ai_alexithymia.md | published: 2023-10-16 - [Ongoing Moving Sale - Text For Info](https://raw.works/ongoing-moving-sale-text-for-info/): content/works/movingsaletextforinfo.md | published: 2023-10-15 - [Generous Residue](https://raw.works/generous-residue/): content/works/GenerousResidue.md | published: 2023-10-14 - [Books Are So 20th Century](https://raw.works/books-are-so-20th-century/): content/works/booksareso20thcentury.md | published: 2023-10-13 - [Squeezed By The Universe](https://raw.works/squeezed-by-the-universe/): content/works/squeezedbytheuniverse.md | published: 2023-10-12 - [6 Hats Helper: Your New Thinking Buddy](https://raw.works/6-hats-helper-your-new-thinking-buddy/): content/works/6hatshelper.md | published: 2023-10-11 - [How To Train ChatGPT On A Book In 5 Minutes](https://raw.works/how-to-train-chatgpt-on-a-book-in-5-minutes/): content/works/howtotrainchatgptonabookin5minutes/index.md | published: 2023-10-10 - [Crickets](https://raw.works/crickets/): content/works/crickets.md | published: 2023-10-09 - [(Some) Questions For a New Project](https://raw.works/some-questions-for-a-new-project/): content/works/somequestionsforanewproject.md | published: 2023-10-08 - [Purple Impressions](https://raw.works/purple-impressions/): content/works/purpleimpressions.md | published: 2023-10-07 - [When The Magic Is Gone](https://raw.works/when-the-magic-is-gone/): content/works/whenthemagicisgone.md | published: 2023-10-06 - [Indra's Web: Part 1](https://raw.works/indras-web-part-1/): content/works/indraswebpart1.md | published: 2023-10-05 - [DALL·E 3 Can Almost Spell](https://raw.works/dalle-3-can-almost-spell/): content/works/dalle3_can_almost_spell/index.md | published: 2023-10-04 - [Plumber.AI](https://raw.works/plumber.ai/): content/works/plumber_ai.md | published: 2023-10-03 - [Knowing About](https://raw.works/knowing-about/): content/works/oct2.md | published: 2023-10-02 - [ Professional 'Advisors' ](https://raw.works/professional-advisors/): content/works/professionaladvisors.md | published: 2023-10-01 - [Why Does Handwriting OCR Suck in 2023?](https://raw.works/why-does-handwriting-ocr-suck-in-2023/): content/works/handwritingOCR.md | published: 2023-09-30 - [Stories <> Data](https://raw.works/stories-data/): content/works/stories_data.md | published: 2023-09-29 - [Don't Get X'd](https://raw.works/dont-get-xd/): content/works/dontgetxd.md | published: 2023-09-28 - [Introducing Rai, my AI Librarian](https://raw.works/introducing-rai-my-ai-librarian/): content/works/introducingRai.md | published: 2023-09-27 - [What If The Web Worked For You?](https://raw.works/what-if-the-web-worked-for-you/): content/works/whatifthewebworkedforyou.md | published: 2023-09-26 - [CO2 Offsets for Walking? Why Not?](https://raw.works/co2-offsets-for-walking-why-not/): content/works/treecard.md | published: 2023-09-25 - [A Recipe for Forgiveness](https://raw.works/a-recipe-for-forgiveness/): content/works/arecipeforforgiveness.md | published: 2023-09-24 - [A Joyous Equinox](https://raw.works/a-joyous-equinox/): content/works/ajoyousequinox.md | published: 2023-09-23 - [Balancing Quantity and Quality in Content Creation](https://raw.works/balancing-quantity-and-quality-in-content-creation/): content/works/qualityandquantity.md | published: 2023-09-22 - [Sharecropping](https://raw.works/sharecropping/): content/works/sharecropping.md | published: 2023-09-21 - [My First Executive AI Win](https://raw.works/my-first-executive-ai-win/): content/works/myfirstexecutiveAIwin.md | published: 2023-09-20 - [Advice Under Uncertainty](https://raw.works/advice-under-uncertainty/): content/works/adviceunderuncertainty.md | published: 2023-09-19 - [Justifying Our Own Existence](https://raw.works/justifying-our-own-existence/): content/works/justifyingourownexistence.md | published: 2023-09-18 - [Automating Social Media Posts for Hugo Sites](https://raw.works/automating-social-media-posts-for-hugo-sites/): content/works/automatingsocialforhugo/index.md | published: 2023-09-17 - [Quantity Over Quality](https://raw.works/quantity-over-quality/): content/works/quantityoverquality.md | published: 2023-09-16 - [Venture Games](https://raw.works/venture-games/): content/works/venturegames.md | published: 2023-09-15 - [There Will Be Blood - Pt. 1](https://raw.works/there-will-be-blood-pt.-1/): content/works/therewillbeblood-pt1.md | published: 2023-09-14 - [Cosmic Reassurance](https://raw.works/cosmic-reassurance/): content/works/fearnotloss.md | published: 2023-09-13 - [Xanadu Now](https://raw.works/xanadu-now/): content/works/xanadu now.md | published: 2023-09-12 - [Paper in Your Pocket](https://raw.works/paper-in-your-pocket/): content/works/Paper in Your Pocket.md | published: 2023-09-11 - [On Feedback](https://raw.works/on-feedback/): content/works/onfeedback.md | published: 2023-09-10 - [Blue Moon](https://raw.works/blue-moon/): content/works/bluemoon.md | published: 2023-09-09 - [Understanding Stories](https://raw.works/understanding-stories/): content/works/understandingstories.md | published: 2023-09-08 - [Chief No Officer](https://raw.works/chief-no-officer/): content/works/chiefnoofficer.md | published: 2023-09-07 - [Mixed Messages](https://raw.works/mixed-messages/): content/works/mixedmessages/index.md | published: 2023-09-06 - [Not planned, intended](https://raw.works/not-planned-intended/): content/works/notplannedintended.md | published: 2023-09-05 - [Multiple Choice: Moments](https://raw.works/multiple-choice-moments/): content/works/momentsmultiplechoice.md | published: 2023-09-04 - [AI doesn't blink](https://raw.works/ai-doesnt-blink/): content/works/AIdoesntblink.md | published: 2023-09-03 - [Where do stories come from?](https://raw.works/where-do-stories-come-from/): content/works/wheredostoriescomefrom.md | published: 2023-09-02 - [Superpowers](https://raw.works/superpowers/): content/works/superpowers.md | published: 2023-09-01 - [A child's nightmare](https://raw.works/a-childs-nightmare/): content/works/childsnightmare.md | published: 2023-08-31 - [Job with a capital J](https://raw.works/job-with-a-capital-j/): content/works/jobwithacapitalj.md | published: 2023-08-30 - [Garbage Truck Mantra](https://raw.works/garbage-truck-mantra/): content/works/garbagetrucks.md | published: 2023-08-29 - [Whose problem?](https://raw.works/whose-problem/): content/works/whoseproblem.md | published: 2023-08-28 - [Bliss & Rent](https://raw.works/bliss-rent/): content/works/bliss and rent.md | published: 2023-08-27 - [Quotes](https://raw.works/quotes/): content/quotes.md | published: undated - [Sign up error](https://raw.works/subscribe-fail/): content/subscribe-fail.md | published: undated - [Thanks for signing up!](https://raw.works/subscribe-success/): content/subscribe-success.md | published: undated ## Documents ### Machine Studying - The Library and The Lab - URL: https://raw.works/machine-studying-the-library-and-the-lab/ - Source: content/works/machine-studying-the-library-and-the-lab.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-06-20T00:00:00Z", "lastmod": "2026-06-20T00:00:00Z", "publishdate": "2026-06-20T00:00:00Z", "title": "Machine Studying - The Library and The Lab" } ``` Content: As usual, I can't help but dive right into anything that comes out of Omar Khattab's lab. "[Machine Studying](https://jacobxli.com/blog/2026/machine-studying/)" is a very interesting blog post that attempts to define "expertise" and "intelligence" based on an agent's ability to quickly learn new material from a corpus. > "In a way, Machine Learning asks how a system can improve from data when we have a precise objective to optimize. **Machine Studying asks what an agent should do when it’s given a declarative corpus and no downstream task.**" I was curious to see what would happen with a modern harness (Codex CLI) and a top model (GPT 5.5 xhigh reasoning). I was able to get a very high score right off the bat (76%, roughly 3x the highest reported in the original blog), which would suggest that either this combination of model and harness has high *expertise* on these tasks (already knows them well), or generally has fairly high *intelligence* for this category of task (ability to gain expertise). Both of the examples in the current version of StudyBench involve testing an agent on its ability to use a specific version of a programming tool (DSPy & OpenClaw). The dates of the version in the test are from March and April of 2026, so I thought that it might be possible that these versions might already be in the training data of GPT 5.5, although [the official page from OpenAI](https://developers.openai.com/api/docs/models/gpt-5.5) says the knowledge cutoff is Dec 1, 2025. This led me to the idea that there are at least two different environments to study in: "the library" and "the lab". In the library - you have books (i.e. documentation) and in the lab you have instruments (i.e. packages). My initial environments for DSPy and OpenClaw gave Codex access to both the documentation and the package. So I asked my buddy Codex (GPT 5.5 xhigh) to build out the remaining 3 environments. ("Lab only" turned out to be way trickier than anticipated as you have to block the package's self-documentation.) So now for each OpenClaw and DSPy StudyBench questions we end up with these four environments: - **Closed book**: the agent only sees the question. It gets no docs, no source tree, no package metadata, and no runtime to experiment with. - **Lab only**: the agent gets no docs, no source tree, and no static study corpus. It can only learn by running small probes against the pinned tool/runtime through the allowed lab surface. - **Library only**: the agent can read and search the official static material: pinned source tree, docs, package metadata, and exposed corpus files. It cannot run the package, import it, execute tests, or use a binary/runtime. - **Library + lab**: the agent gets both surfaces. It can read the official static material and also run experiments against the pinned environment. | Treatment | Library | Lab | DSPy | OpenClaw | Mean | | --- | --- | --- | ---: | ---: | ---: | | Closed book | no | no | 51.70 | 6.35 | 33.56 | | Lab only | no | yes | 78.07 | 9.70 | 50.72 | | Library only | yes | no | 82.97 | 60.30 | 73.90 | | Library + lab | yes | yes | 85.40 | 62.05 | 76.06 | So these results suggest a few things: - (Obviously) having access to both the documentation and the tool itself (library and lab) allows the agent to acquire the most expertise on DSPy and OpenClaw. - GPT 5.5 appears to have much better built-in expertise (closed book score) about DSPy than OpenClaw. This makes sense, given that DSPy is many years old, and that OpenClaw is both very new and also has had three different names in its brief history. - "Lab only" (tools but no documentation) led to modest improvements for both environments. "Library only" (docs and source code but no runnable package available) was slightly higher than "lab only" for DSPy, but dramatically higher for OpenClaw. I haven't done an in-depth analysis of this, but again my suspicion is that it is related to the newness of OpenClaw, the quality of the documentation, and the structure of the StudyBench questions. - Today's SOTA coding agents may already be highly effective learners, when given access to documentation, source code, and runnable packages. I did not make any attempts to control the budget (time, tokens, turns) or to vary the reasoning level (fixed xhigh) - so I cannot compute either the "expertise" or the "intelligence" in the framing of the Machine Studying post. These are obvious next steps, in addition to trying out some different models and harnesses. My environments and results are available at: [github.com/rawwerks/studybench-lab-and-library](https://github.com/rawwerks/studybench-lab-and-library) Until next time...study hard! ### Engineering Skills - URL: https://raw.works/engineering-skills/ - Source: content/works/engineering-skills.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-05-12T00:00:00Z", "image": "/images/engineering-skills-og.png", "lastmod": "2026-05-12T00:00:00Z", "publishdate": "2026-05-12T00:00:00Z", "title": "Engineering Skills" } ``` Content: Agent skills are a powerful and portable way to transform generally capable AI agents into specifically useful tools — even teammates. There are skills for almost anything you can imagine: [doing your taxes](https://www.jeffreyemanuel.com/writing/tax_gpt_using_ai_for_tax_prep), [setting up Cloudflare](https://github.com/cloudflare/skills), [speaking like a caveman](https://skills.sh/juliusbrussee/caveman/caveman), [generating algorithmic art](https://skills.sh/anthropics/skills/algorithmic-art)...there's even a skill dedicated to [industrial brutalism](https://skills.sh/leonxlnx/taste-skill/industrial-brutalist-ui). Skills became so easy to make that the next challenge became *finding* them. This was quickly solved by various "skill registries" like [Vercel's skills.sh](https://skills.sh/). [OpenClaw](https://github.com/openclaw/openclaw) was a huge inflection point for the skills boom — it made agent skills first-class in the product architecture and provided a high-permission substrate for millions of people to experiment with. Of course, then safety became a problem. That's a topic for a longer post, but in case you are curious, I developed the [Skill Safety Data Sheet](https://skillsafetydatasheet.org/) as an analogy to material safety data sheets — for evaluating the risks of specific agent skills. ## So where do skills stand today? If you are an information hoarder like me, your computer is also full of dozens or even hundreds of awesome agent skills — some that 10x devs shared on GitHub, and some that Claude Code made for you after you got tired of saying the same thing over and over and over again. And if you are like me, your Claude Code and Codex have a terrible habit of finding random skills you don't even remember installing. Worse, they never load the ones you *just* added — or at least not until you scold them. There are two details of the skills implementation that make it very unreliable: 1. **LLMs are nondeterministic.** They aren't guaranteed to load the right skill at the right time. 2. **Progressive disclosure** (implemented as a context window workaround) means the agent has to go looking for the information fresh each time. The clanker has to *think* to find the skill first, which is really inefficient. In fact, [Vercel](https://vercel.com/blog/agents-md-outperforms-skills-in-our-agent-evals) recently showed that **model-mediated skill activation lost to a simple index** — a compressed 8KB `AGENTS.md` hit a 100% pass rate on their Next.js evals while a carefully crafted skill maxed out at 79%, and the skill was never invoked at all in 56% of cases. The winning approach still used a form of progressive disclosure — it just moved the routing layer into stable passive context. (In case you're curious, I made [`dirpack`](https://github.com/rawwerks/dirpack) as a general utility for creating indices of a fixed token budget for any directory.) ## Engineer the skills! So how can we take advantage of the power of agent skills without giving up our own agency to decide when and how they should be used? **Engineer the skills!** Right now I'm having a lot of fun working on [OpenProse](https://github.com/openprose/prose) — a "programming language" that is compiled inside of a coding agent. If this sounds sci-fi, it is...and OpenProse only really works with today's top models. The fun thing about OpenProse is that you can express very complex workflows in very simple markdown files. As an example, using the legacy v0 syntax purely for brevity: ```openprose input topic loop until **editor approves** (max: 5): session "research {{topic}}, address editor's prior notes" session "draft from research, revise per prior notes" session "review draft: approve as report or emit notes" return report ``` This kind of logical statement is impossible to express in any other language. Prose is super fun! But the problem I quickly ran into is that many of the prose programs I would want to run assume that my agents will use a *specific* skill. So I recently added the ability to deterministically declare agent skills inside prose programs — [here's the PR](https://github.com/openprose/prose/pull/62). As a fun example of what's possible with this new feature, I created **[auto-pocock](https://github.com/openprose/prose/tree/main/skills/open-prose/examples/auto-pocock)**: a headless prose program that incepts your favorite coding agent into running a deterministic sequence of [Matt Pocock's engineering skills](https://github.com/mattpocock/skills) (`grill-with-docs` → `to-prd` → `to-issues` → `tdd` → verify → commit), all from one input — a description of the feature you want built. {{< twitter-center user="raw_works" id="2054246083849507143" >}} This combination of specific instructions (**skills**), deterministic processes (**the prose contract**), and nondeterministic magic (**coding agents**) is extremely versatile. By engineering skills with OpenProse, you can express complex multi-agent workflows, imbue each agent with detailed discipline, and hopefully get a much-needed break from the keyboard. ### Code Execution as Reasoning - URL: https://raw.works/code-execution-as-reasoning/ - Source: content/works/code-execution-as-reasoning.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-05-11T00:00:00Z", "image": "/images/code-execution-as-reasoning-og.png", "lastmod": "2026-05-11T00:00:00Z", "publishdate": "2026-05-11T00:00:00Z", "title": "Code Execution as Reasoning" } ``` Content: I personally do not care if my AI programs do their reasoning in latent space or code. I want results. I am currently very intrigued by [LongCoT](https://longcot.ai), a new benchmark that is designed to push the limits of what is possible with today's LLMs. Part of the original intention of the benchmark was to create something that would be both challenging for LLMs and less enmeshed with the details of the harness. My [recent results using DSPy.RLM]({{< ref "rlms-are-sota-on-longcot.md" >}}) have caused a bit of drama with some of the leaderboard owners, and the creation of a new tools-enhanced leaderboard. I understand the academic value in having "pure latent space" results without tools, but it just isn't interesting to me...I want my agents to have tools. So I set out to give the LLMs their [desire path](https://steve-yegge.medium.com/software-survival-3-0-97a2a6255f7b) - the python tools that they tried to use in my prior benchmarking experiments. This works surprisingly well with DSPy.RLM and Opus 4.7, which achieved a new SOTA on LongCoT-mini. **Opus 4.7 + DSPy.RLM → 377/500 (75.4%)** on LongCoT-Mini — new top of the leaderboard, and a clear jump over the [Sonnet 4.5 + DSPy.RLM 45.4% I posted in April]({{< ref "rlm-longcot-sonnet.md" >}}). | Mini | Opus 4.7 + RLM | Sonnet 4.5 + RLM | Sonnet 4.5 vanilla | |---|---|---|---| | chess | 98/98 | 85/100 | 0/100 | | logic | 101/101 | 106/110 | 0/110 | | chemistry | 66/98 | 31/100 | 13/100 | | cs | 71/97 | 4/100 | 0/100 | | math | 41/77 | 6/95 | 0/95 | | **total (official /500)** | **377/500 (75.4%)** | 227/500 (45.4%) | 13/500 (2.6%) | The new runs scored on the 471-question working set after the LongCoT team audited out 29 Mini questions as unsolvable. The official /500 totals here count those audited-out rows as wrong, matching the denominator used for the Sonnet baselines. [The paper](https://arxiv.org/abs/2604.14140) splits the five domains into two classes: *implicit* (Logic, Chess, CS), where the dependency structure can be externalised to code, and *explicit compositional* (Math, Chemistry), where it can't. The headline claim is that even with code execution enabled, RLM lifts the implicit class but leaves the compositional class near zero — direct quote: *"explicit compositional domains (Math, Chemistry) remain at zero."* My [April Sonnet run]({{< ref "rlm-longcot-sonnet.md" >}}) replicated that shape (math 6/95, hardest cs templates 0/75). Opus + RLM gets math up to 41/77 and cs to 71/97. Not zero. Special thanks to [Prime Intellect](https://www.primeintellect.ai/) for sponsoring inference on this experiment — I promise I will publish my LongCoT environments soon. Even with that support I ran out of credits quickly, so I switched to OpenAI Codex CLI on the latest GPT-5.5 at "xhigh" reasoning, to put my $200/mo sub to work. **Codex CLI + gpt-5.5 "xhigh" → 398/500 (79.6%)** on LongCoT-Mini — +21 rows over Opus. | Mini | Codex 5.5 xhigh | Opus 4.7 + RLM | |---|---|---| | chess | 98/98 | 98/98 | | logic | 100/101 | 101/101 | | chemistry | 78/98 | 66/98 | | cs | 67/97 | 71/97 | | math | 55/77 | 41/77 | | **total (official /500)** | **398/500 (79.6%)** | 377/500 (75.4%) | Two scaffolds, two models, similar bottom line. Codex's persistent in-sandbox Python loop is grinding harder on chemistry and math, while DSPy.RLM holds a small cs edge. The LongCoT-Mini scoreboard is starting to feel more like a measure of how the agent's tool loop is wired than of which frontier endpoint is behind it. Mini is the *easy* slice. The full benchmark — medium + hard, ~2000 questions, where the dependency DAGs grow long enough to actually bite — is the real test of the paper's compositional-walls claim. Over the weekend I let Codex loose on it; this time it took multiple Codex subscriptions to finish. **Codex CLI gpt-5.5 xhigh on full LongCoT: 1446/1995 (72.5%)** — about **3× the top of the live [longcot.ai](https://longcot.ai) Open Harness leaderboard**, where the LongCoT team's own GPT 5.2 + rlm run holds #1 at 25.12%. ([My April Qwen 3.5 27B + DSPy.RLM run](/rlms-are-sota-on-longcot/) is #2 at 22.18%.) | Full LongCoT · Codex 5.5 xhigh | class | medium | hard | total | Open-Harness #1 (GPT 5.2 + rlm) | |---|---|---|---|---|---| | logic | implicit | 187/195 (95.9%) | 165/199 (82.9%) | **89.3%** | 68.3% | | cs | implicit | 150/150 (100.0%) | 210/250 (84.0%) | **90.0%** | 26.7% | | chess | implicit | 92/150 (61.3%) | 200/250 (80.0%) | **73.0%** | 30.6% | | math | compositional | 110/150 (73.3%) | 168/250 (67.2%) | **69.5%** | 0.0% | | chemistry | compositional | 114/200 (57.0%) | 50/200 (25.0%) | **41.0%** | 0.0% | | **total** | | **77.3%** | **69.0%** | **72.5%** | **25.12%** | The compositional class — Math and Chemistry, the domains the paper said would *"remain at zero"* even with code execution — comes in at **69.5%** and **41.0%**. cs **medium goes 150/150**. The walls aren't walls; with a stronger model and a more aggressive tool loop, they're just programs. With the full LongCoT now in hand, I think we can clearly state that the paper's compositional walls don't hold. Math and Chemistry — the domains the paper claimed would *"remain at zero"* even with code execution — come in at 69.5% and 41.0%. The wall isn't compositional reasoning; it's the harness used to measure it. I think there is very clear evidence of something many people have been saying for the past ~18 months: *it's not (just) the model, it's the harness*. Additionally, it is very clear proof that "coding agents" are useful for long horizon tasks that don't necessarily present themselves as coding problems. ### RLMs are the new reasoning models - URL: https://raw.works/rlms-are-the-new-reasoning-models/ - Source: content/works/rlms-are-the-new-reasoning-models.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-04-20T00:00:00Z", "image": "/images/rlm-lobster-fractals.jpg", "lastmod": "2026-04-20T00:00:00Z", "publishdate": "2026-04-20T00:00:00Z", "title": "RLMs are the new reasoning models" } ``` Content: Reasoning models were the first clear proof that language model capability can scale with test-time compute. [Recursive language models (RLMs)](https://arxiv.org/abs/2512.24601) ask what the correct abstraction for spending that compute is. The insight behind RLMs is obvious in hindsight: it is the direct marriage of two important axes of model capability — reasoning and tool use. This is more radical than it first sounds. RLMs collapse reasoning and tool use into a single inference abstraction: the model treats its own prompt as an environment it can inspect, slice, and recursively query. *Context itself becomes the object of computation.* This post is my attempt to explain why RLMs matter. I define what a RLM actually is, place it in the short history of reasoning and tool use, walk through the ~6 months of empirical results that have quietly turned "RLM" from a benchmark trick into the next reasoning paradigm, flag the honest limitations, and point at a few places to start building. ## What is a RLM? A Recursive Language Model, as introduced by [Zhang, Kraska, and Khattab](https://arxiv.org/abs/2512.24601), is an inference paradigm in which a language model treats its input prompt as an environment rather than a fixed string. The root LM is given a REPL in which the prompt is bound to a variable it can inspect, slice, and partition programmatically. When it decides a region is worth a closer look, it issues a recursive subcall — to itself or another LM — over that slice and incorporates the result. Recursion bottoms out at the base model's ordinary forward pass. One consequence is that input size is no longer a hard ceiling on the computation. The paper reports RLMs processing inputs up to two orders of magnitude beyond the underlying model's context window and outperforming vanilla frontier LLMs and common long-context scaffolds across four long-context tasks. Beyond long-context answering, recent results demonstrate that RLMs are a powerful paradigm for a wide variety of challenging tasks. ## Reasoning & Tool Use — A Brief History Reasoning and tool use are related, but they are not the same thing. Reasoning is about how well a model can allocate inference-time compute to a problem: break it down, explore alternatives, verify intermediate steps, backtrack, and choose a better answer. Early reasoning gains came from methods like chain-of-thought, self-consistency, and later tree-search-style prompting. Those methods improve how the model thinks even when it never touches the outside world. Tool use is about whether a model can decide to call an external function, search engine, calculator, browser, code runner, or UI action; pass the right arguments; interpret the result; and continue. That is partly a reasoning problem, but it is also an interface and reliability problem: schemas, argument formatting, retries, stop conditions, state tracking, and error recovery. Toolformer made this distinction especially clear by treating tool use as something a model could learn during generation. Historically, the timeline looks roughly like this: **2022: reasoning first, mostly without tools.** Chain-of-thought prompting showed that asking models to generate intermediate reasoning steps could dramatically improve multi-step reasoning. Self-consistency pushed this further by sampling multiple reasoning paths and selecting the most consistent answer. The key lesson was that a large share of “reasoning” gains could come from spending more inference-time compute on the same prompt, not just from adding more knowledge. - Chain-of-Thought Prompting Elicits Reasoning in Large Language Models — - Self-Consistency Improves Chain of Thought Reasoning in Language Models — **Late 2022: the first real bridge between reasoning and acting.** ReAct was the key milestone. It framed the model as alternating between reasoning traces and external actions such as retrieval or environment interaction. This was the moment the field started to see tool use not as a one-off API call, but as a loop in which reasoning selects actions and tool outputs reshape the next reasoning step. - ReAct: Synergizing Reasoning and Acting in Language Models — **2023: tool use becomes an API discipline, not just a prompting trick.** Toolformer argued that models could learn when to call tools, which tools to call, and how to incorporate the results. Around the same time, vendors began standardizing function-calling interfaces. OpenAI’s June 2023 function calling release was a major product milestone because it made structured tool invocation reliable enough for developers to build on. This improved tool-use reliability faster than it improved deep reasoning. - Toolformer: Language Models Can Teach Themselves to Use Tools — - OpenAI, “Function calling and other API updates” — **2023 also deepened the separation between reasoning and tool use.** Tree of Thoughts made it even clearer that inference-time reasoning could improve through internal search alone. It let models explore multiple candidate thought branches, look ahead, and backtrack. That is search over reasoning traces. It can be paired with tools, but it does not require them. - Tree of Thoughts: Deliberate Problem Solving with Large Language Models — **2024: reasoning models become their own product category.** OpenAI’s o1 launch was the clearest signal. The company described o1 as a model family designed to “spend more time thinking before they respond,” and the initial API announcement explicitly noted that features like function calling were not yet included. That was strong evidence that, product-wise, reasoning and tool use were still separable. - OpenAI, “Introducing OpenAI o1-preview” — - OpenAI, “Introducing OpenAI o1” — **2024 is also when agentic tool use got much more serious.** Anthropic’s Claude 3.5 Sonnet emphasized stronger tool use for coding and agentic tasks, and later in 2024 Anthropic introduced computer use: a model interacting with a real computer via screenshots, mouse, and keyboard. This is a good example of the two axes starting to merge into one agentic stack. - Anthropic, “Introducing computer use, a new Claude 3.5 Sonnet, and Claude 3.5 Haiku” — - Anthropic, “Developing a computer use model” — **Late 2024 into 2025: vendors start presenting tool use as native, but still distinct from thinking.** Google’s Gemini 2.0 messaging explicitly framed the model family around the “agentic era” and native tool use, while keeping “thinking” as a distinct capability for harder multi-step planning. That split mirrors the real architecture: one layer governs deliberation, another governs interaction with external affordances. - Google, “Google Gemini AI update, December 2024” — RLMs are the abstraction where that split finally collapses. The past ~6 months of results are what make the case concrete. ## Recent RLM Results The arc of RLM results moves through three successive failure modes of the single forward pass: **long context**, then **memory**, then **long reasoning**. Each has been demonstrated by its own benchmark — [Oolong](https://arxiv.org/abs/2511.02817), [LongMemEval](https://arxiv.org/abs/2410.10813), and [LongCoT](https://arxiv.org/abs/2604.14140) respectively — and RLM-style systems have posted leading numbers on all three. Just as importantly, the follow-up work is already splitting into two camps: work that strengthens the original RLM implementation, and work that argues the deeper win is broader **externalized program search** rather than recursion alone. Part of what makes RLMs challenging to appreciate is that frankly there aren't very many benchmarks that really showcase the differences. In particular, I don't view Oolong and LongMemEval as having much correlation to performance on real world agentic tasks. LongCoT is much more exciting to me, but it is brand new and only time will tell how it holds up. **2024: the memory target appears.** [LongMemEval](https://arxiv.org/abs/2410.10813) defines the benchmark for long-term interactive memory: 500 questions over sustained chat histories spanning extraction, multi-session reasoning, temporal reasoning, knowledge updates, and abstention. It matters here because it gives RLM-style systems a way to test whether recursive/tool-mediated processing can function as a *memory system*, not just a long-context hack. - [LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://arxiv.org/abs/2410.10813) **October 2025: the original public RLM write-up lands.** In [Recursive Language Models](https://alexzhang13.github.io/blog/2025/rlm/), Alex Zhang introduces the core idea: treat the prompt as an external environment, manipulate it through a REPL, and recursively subquery models over slices of context. The post reports an unusually strong early result profile: a GPT-5-mini RLM beats GPT-5 by more than 2× on an Oolong split while being cheaper per query on average, beats ReAct + test-time indexing/retrieval on a BrowseComp-Plus-derived long-context research task, and does not visibly degrade even at 10M+ input tokens. - [Recursive Language Models (original blog post)](https://alexzhang13.github.io/blog/2025/rlm/) **November 2025: Oolong raises the bar for long-context reasoning.** [Oolong](https://arxiv.org/abs/2511.02817) is important because it measures something harder than needle-in-a-haystack retrieval: models have to analyze many local chunks and then aggregate them into a global answer. At release, GPT-5, Claude-Sonnet-4, and Gemini-2.5-Pro all score under 50% on both splits at 128K, making Oolong the clearest early benchmark for the kind of “context as workspace” reasoning RLM is trying to solve. - [Oolong: Evaluating Long Context Reasoning and Aggregation Capabilities](https://arxiv.org/abs/2511.02817) **December 2025: the arXiv paper formalizes RLM.** The [Recursive Language Models paper](https://arxiv.org/abs/2512.24601) turns the blog’s intuition into a general inference paradigm: prompts are externalized, the LM programmatically inspects and partitions them, and recursive subcalls become part of test-time compute. The headline results are strong: RLMs process inputs up to two orders of magnitude beyond model context windows, outperform vanilla frontier LLMs and common long-context scaffolds across four long-context tasks at comparable cost, and a fine-tuned RLM-Qwen3-8B improves 28.3% on average over its base model. - [Recursive Language Models (arXiv, Dec. 2025; revised Jan. 2026)](https://arxiv.org/abs/2512.24601) **February 2026: RLM starts posting real “memory” numbers.** In [Recursive Language Models as Memory Systems](https://raw.works/recursive-language-models-as-memory-systems/), I reported early LongMemEval results with DSPy.RLM: 87.2% for a baseline Gemini 3 Flash setup, 89.2% with tools + a delegation prompt, and 89.8% with an observational-memory-style structured scaffold. That was a public Top-5-ish result at the time, below Mastra’s 94.87% but already strong evidence that RLM can act as a competitive memory system without a classical retrieval stack. In [ypi: a recursive coding agent](https://raw.works/ypi-a-recursive-coding-agent/), I show an earlier tool-use REPL path scoring 77.6% on LongMemEval — a useful datapoint because it shows the gradient from “tool-using agent” to “true recursive scaffold” inside the same implementation lineage. - [Recursive Language Models as Memory Systems](https://raw.works/recursive-language-models-as-memory-systems/) - [ypi: a recursive coding agent](https://raw.works/ypi-a-recursive-coding-agent/) - [Observational Memory: 95% on LongMemEval](https://mastra.ai/research/observational-memory) **March 2026: follow-up papers clarify both the strengths and the limits.** [Think, But Don’t Overthink](https://arxiv.org/abs/2603.02615) reproduces RLM and finds that depth-1 recursion helps on Oolong, but deeper recursion can “overthink,” hurting accuracy and exploding runtime and token cost. [Recursive Language Models Meet Uncertainty](https://arxiv.org/abs/2603.15653) pushes a sharper critique: recursion itself is not the whole secret, and uncertainty-aware self-reflective program search can improve up to 22% over RLM under the same time budget. Then [Coding Agents are Effective Long-Context Processors](https://arxiv.org/abs/2603.20432) generalizes the broader thesis: off-the-shelf coding agents outperform published SOTA by 17.3% on average, and on Oolong-Synthetic / Oolong-Real their reported scores (71.75 / 33.73) exceed the paper’s RLM baselines (64.38 / 23.07). That does not really refute RLM; it suggests RLM was the first clearly articulated expression of a larger family of executable, tool-mediated long-context reasoning systems. - [Think, But Don't Overthink: Reproducing Recursive Language Models](https://arxiv.org/abs/2603.02615) - [Recursive Language Models Meet Uncertainty: The Surprising Effectiveness of Self-Reflective Program Search for Long Context](https://arxiv.org/abs/2603.15653) - [Coding Agents are Effective Long-Context Processors](https://arxiv.org/abs/2603.20432) **April 2026: the theory catches up to the results.** In [The Mismanaged Geniuses Hypothesis](https://alexzhang13.github.io/blog/2026/mgh/), Zhang reframes the whole arc: RLM is not just a benchmark trick for long prompts, but a more expressive scaffold for plans written through code execution, recursive subcalls, and tools-as-functions. That is a useful conceptual update because it connects the empirical results back to the bigger claim: reasoning performance is starting to look less like a property of a single forward pass and more like a property of how well a model can manage executable external computation. - [The Mismanaged Geniuses Hypothesis](https://alexzhang13.github.io/blog/2026/mgh/) The empirical case moves just as quickly. **April 2026: the benchmark story shifts from long context to long reasoning.** [LongCoT](https://arxiv.org/abs/2604.14140) introduces 2,500 expert-designed problems for long-horizon chain-of-thought reasoning. At release, the best published models are still under 10% accuracy (GPT-5.2 at 9.8%, Gemini 3 Pro at 6.1%), which makes it an ideal test for whether recursive scaffolds are merely “good at reading long context” or whether they genuinely unlock *reasoning depth*. - [LongCoT: Benchmarking Long-Horizon Chain-of-Thought Reasoning](https://arxiv.org/abs/2604.14140) **April 2026: RLM immediately breaks LongCoT open.** In [LongCoT — A benchmark worthy of a RLM’s attention](https://raw.works/longcot-a-benchmark-worthy-of-a-rlms-attention/), I showed Claude Sonnet 4.5 + DSPy.RLM reaching **45.4%** on LongCoT-Mini versus **2.6%** for the same model without recursion/tools. Then in [RLMs are SOTA on LongCoT](https://raw.works/rlms-are-sota-on-longcot/), I show the scaffold doing almost all of the lifting for small open models: Qwen3-8B jumps from **0/507** to **33/507 (6.5%)** on LongCoT-Mini; Qwen3.5-9B + DSPy.RLM reaches **15.69%** on full LongCoT, about **1.6×** GPT-5.2 on the same slice; and Qwen3.5-27B + DSPy.RLM reaches **22.18%**, more than **2×** GPT-5.2. If these numbers hold up, they are some of the clearest evidence yet that recursive scaffolds can manufacture reasoning performance that is not visible in the base model alone. - [LongCoT — A benchmark worthy of a RLM's attention](https://raw.works/longcot-a-benchmark-worthy-of-a-rlms-attention/) - [RLMs are SOTA on LongCoT](https://raw.works/rlms-are-sota-on-longcot/) The arc is now hard to ignore. Oolong gives the **long-context** failure mode. LongMemEval gives the **memory** version. LongCoT gives the **long-reasoning** version. Across all three, the recurring pattern is the same: when the task requires navigating, decomposing, and aggregating information over a structure that is too large or too entangled for one passive forward pass, recursive tool-mediated processing starts to look less like an implementation trick and more like the next reasoning paradigm. ## Challenges with RLMs A new paradigm is not a clean paradigm. Reasoning models were scoffed at for being too expensive. Early tool calling reliability was horrible. Even some leading reasoning models today are pretty bad at function calling. RLMs have their challenges. My earliest contributions to the standalone RLM package and the DSPy.RLM implementation were purely practical: budgets, timeouts, managing the recursion depth. Recursion sounds cool but isn't always a good thing. Remember those viruses that would make your browser open a million popups until your computer crashed? Recursion can be scary. [![Mario Zechner (@badlogicgames): "i'm scared" — in reply to ypi](/images/mario-zechner-im-scared.png)](https://x.com/badlogicgames/status/2022071738146988076) The most obvious limitations right now are cost and time. RLMs are expensive. They can take a long time. Worse, in the naive implementation that time is unpredictable and unbounded, because the model is deciding for itself how to decompose the problem. Cost and time will be solved. Use smaller or faster models for each sub-call, and balance the agent-native "self-similar" decomposition with deterministic control of the graph topology and timeline. The harder challenge, or at least the challenge that is more interesting to me personally, is how to get the language models to "act recursively". Obviously the concepts of recursion are in the pre-training data. Clearly reasoning and parallel tool calling are behaviors that the post-training incentivizes. Sub-agents are arguably a close behavioral analog to RLMs. And yet anyone who has worked with RLMs will tell you that the models generally suck at behaving recursively. It is not in their nature to decompose their prompt into sub-queries for many other instances of themselves to help solve them. ## What's next? Well one obvious next step is to explicitly post-train the models in a RLM harness. Alex Zhang et al. are actively working in this area: [MIT OASYS on HuggingFace](https://huggingface.co/mit-oasys) (see e.g. [`mit-oasys/rlm-qwen3-8b-v0.1`](https://huggingface.co/mit-oasys/rlm-qwen3-8b-v0.1)). But what is the reward function for "optimal recursion"? I suspect this is a multi-billion-dollar question. The most surprising result to me from my last few days of experimenting was [how well very small models can do in RLM harnesses](https://raw.works/rlms-are-sota-on-longcot/). These models are small enough to run on consumer devices, which potentially means that they offer an opportunity to upset the current "balance of power" between the GPU-rich and GPU-poor. Yes, more money means you can run more. The best GPUs will always be faster. A RLM of Opus is smarter than a RLM of Llama 3. But I cannot help but feel excited and empowered to believe that an individual or consortium running many instances of small models on affordable/legacy/local compute infrastructure can now access model capabilities that are on par with or exceeding those of the most expensive LLMs from the frontier labs. If that is even directionally right, the frontier stops being a place only the largest labs can reach. ## Getting Started with RLMs Here are just a few of the many ways to get started with RLMs: - [**alexzhang13/rlm**](https://github.com/alexzhang13/rlm) — the reference implementation from Alex Zhang and the RLM paper authors; the cleanest place to read the core recursion loop. - [**dspy.RLM**](https://dspy.ai/api/modules/RLM/) — the DSPy integration, which exposes RLM as a composable module inside larger DSPy programs and is what I've been using for most of my own experiments. - [**ax-llm/ax**](https://github.com/ax-llm/ax) — a TypeScript DSPy-style framework with first-class RLM support via `AxAgent`-driven recursive decomposition, bounded sub-queries, and a persistent JS runtime. - [**rawwerks/rlm-cli**](https://github.com/rawwerks/rlm-cli) — my CLI wrapper around `rlm` with directory-as-context, JSON-first output, and self-documenting commands, for running RLMs against local repos and folders. - [**rawwerks/ypi**](https://github.com/rawwerks/ypi) — my recursive coding agent built on [Pi](https://github.com/badlogic/pi-mono): one `rlm_query` tool, one `rlm_map` fanout helper, and per-child `jj` workspaces for isolated recursive execution. ## P.S. I almost forgot: fractals. [![Lobster mandelbulb](/images/lobster-mandelbulb.png)](https://x.com/raw_works/status/2022010444492517469) ### RLMs are SOTA on LongCoT - URL: https://raw.works/rlms-are-sota-on-longcot/ - Source: content/works/rlms-are-sota-on-longcot.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-04-19T00:00:00Z", "lastmod": "2026-04-19T00:00:00Z", "publishdate": "2026-04-19T00:00:00Z", "title": "RLMs are SOTA on LongCoT" } ``` Content: A few days ago I showed [Sonnet 4.5 + `dspy.RLM` hitting 45.4% on LongCoT-Mini](/longcot-a-benchmark-worthy-of-a-rlms-attention/). Exciting results, but a bit pricey for my taste. So I set out to see what might be possible with some very small models. First, I wanted to run a 3x2 comparison matrix of doing LLM vs. [RLM](https://github.com/alexzhang13/rlm) vs. [DSPy.RLM](https://dspy.ai/api/modules/RLM/) for both [Qwen 3 8B](https://huggingface.co/Qwen/Qwen3-8B) and the [MIT OASYS RLM finetune](https://huggingface.co/mit-oasys/rlm-qwen3-8b-v0.1). I will need to save the full analysis for another day (I wasn't really able to get the finetuned model working), but the meaningful result is that on [LongCoT](https://longcot.ai) mini, DSPy.RLM can take Qwen 3 8B Instruct from literally 0/507 correct to 33/507 (6.5%). This would be #7 on the leaderboard, from an 8B model! {{< x user="raw_works" id="2045208764509470742" >}} So my immediate next thought was: "what about Qwen 3.5 9B"? This hit 17.2% on LongCoT mini (3rd place), and was so cheap that I decided to run the full benchmark (all 2500 questions)! (Now using Together AI via OpenRouter, I don't think their endpoint is quantized but I'm not 100% sure.) Surprisingly, DSPy.RLM with Qwen 3.5 9B is comfortably SOTA on the full LongCoT at 15.69%, outdoing GPT 5.2 by ~1.6x. {{< x user="raw_works" id="2045581200622841941" >}} Now I was having too much fun, so I had to run Qwen 3.5 27B (this time via Alibaba Cloud via OpenRouter)...and unsurprisingly, a new LongCoT full king is crowned at 22.18%. {{< x user="raw_works" id="2045818627006279745" >}} I'm really excited to finally have a meaningful benchmark that can clearly demonstrate the power of RLMs. This clearly deserves a much longer writeup, which I hope to post soon! (And I'm now running [Qwen 3.6 35B-A3B](https://huggingface.co/Qwen/Qwen3.6-35B-A3B) at the suggestion of many folks on X.) ### LongCoT - A benchmark worthy of a RLM's attention - URL: https://raw.works/longcot-a-benchmark-worthy-of-a-rlms-attention/ - Source: content/works/rlm-longcot-sonnet.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-04-16T00:00:00Z", "lastmod": "2026-04-16T00:00:00Z", "publishdate": "2026-04-16T00:00:00Z", "title": "LongCoT - A benchmark worthy of a RLM's attention" } ``` Content: After [the LongMemEval experiments in February](https://raw.works/recursive-language-models-as-memory-systems/), I've been hungry to find a better benchmark that will actually showcase the power of recursive language models (RLMs) on useful tasks. [LongCoT](https://longcot.ai) is exactly that: a benchmark built to stress-test long-horizon reasoning. As soon as I saw the benchmark, I aimed [DSPy.RLM](https://dspy.ai/api/modules/RLM/) at it. (Even before [reading the paper](https://arxiv.org/abs/2604.14140).) ## The setup - **Model:** `claude-sonnet-4-5` for both conditions. Same `max_tokens=64000`, same judge models, same prompts. - **RLM:** stock `dspy.RLM` 3.1.3, `max_iterations=50`, default Pyodide REPL, `sub_lm=lm`. - **Vanilla:** raw Anthropic SDK, single user message, no tools. Leaderboard shape. - **Dataset:** LongCoT-Mini, all 500 questions (easy slices across logic / cs / chemistry / chess / math). The entire RLM surface area is one dspy Signature: ```python class LongCoTSolve(dspy.Signature): """Solve a LongCoT problem. The `prompt` already contains the full problem statement and the answer format requirement (always ends with `solution = ...`). Reason through the problem with the available REPL, then return the final response — which MUST contain the literal `solution = ...` line as instructed. """ prompt: str = dspy.InputField(desc="Full LongCoT problem prompt with answer-format instructions") response: str = dspy.OutputField(desc="Full final response containing the required `solution = ...` line") ``` ## The headline | | Vanilla | RLM | |---|---|---| | Correct | 13 / 500 | **227 / 500** | | Accuracy | 2.6% | **45.4%** | | Captured cost | $31 | $621 | On the full 500-row overlap: 219 wrong→right flips, 5 right→wrong, 268 both-wrong, 8 both-right. The vanilla 2.6% lines up with the published Sonnet 4.5 Mini number, so the control is calibrated, not sandbagged. ## Per-task | Task | RLM | Vanilla | |---|---|---| | Dungeon · Packaging · Hanoi · Wizards · TrapezoidCounting · Sudoku | 15/15 each (💯) | 0/15 each | | BlocksWorld | 9/10 | 0/10 | | Sokoban | 7/10 | 0/10 | | Chess | **85/100** | 0/100 | | Chemistry | 31/100 | 13/100 | | cs / DistMem | 4/25 | 0/25 | | cs / MaxFlow-MinCut + Hindley-Milner | **0/75** | 0/75 | | math | **6/95** | 0/95 | The pattern is coherent: RLM crushes anything whose dependency structure externalises cleanly to code. The orchestrator writes a short Python program, the REPL runs it, the answer comes out. Logic puzzles, Hanoi, Sudoku, chess with Pyodide's `chess` module — all 💯 or near it. The walls are the opposite picture. Hindley-Milner and MaxFlow-MinCut go 0/75 because the orchestrator can't find a decomposition where subproblems can be usefully farmed out — exactly the "graph-structured dependencies" failure the paper calls out. And math? The paper's wall holds at least for now, for Sonnet 4.5. 6/95 on Mini isn't zero, but it's terrible. Sonnet 4.5 × dspy.RLM replicates the paper's math result on a different model and split. ## What I think this means for the paper The paper's RLM discussion is genuinely thin — one paragraph, one figure, no dedicated table. With that as the bar, cross-model replication is useful: - **Logic, chess, CS wins: replicate and amplify.** Same shape on a different frontier model. - **Math stays at zero: maybe?** Model swap doesn't rescue it. But it's also from a baseline of 6 so you can argue it's either modest or infinite improvement. - **Chemistry lifts modestly** (13 → 31), which is the only spot where I'd push back on the paper's phrasing — but I'm both a chemist and RLM addict. P.S. - RLMs are expensive, and supposedly the 500-problem mini version is the *easy* subset of the full 2500-problem set. So...who wants to fund the Opus 4.7 run? ### mycelium - an underground information network for agents - URL: https://raw.works/mycelium-an-underground-information-network-for-agents/ - Source: content/works/mycelium.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-03-27T00:00:00Z", "lastmod": "2026-03-27T00:00:00Z", "publishdate": "2026-03-27T00:00:00Z", "tags": [], "title": "mycelium - an underground information network for agents" } ``` Content: Over the last 2 days, we've stumbled upon a really powerful coding agent interaction pattern: git notes as an underground information network. [Git notes](https://git-scm.com/docs/git-notes) are both ubiquitous (part of git) and "invisible" (GitHub chose not to display them). This presents a very interesting communication channel for agents, who can now include rich details and discussions about the code without cluttering up the "visible" layer of the repo. [mycelium](https://github.com/openprose/mycelium) is my tool to make these interactions easier. ```bash # agent arrives, reads what's known about a file mycelium.sh context src/auth.ts # agent works... # agent leaves a note explaining what it did mycelium.sh note HEAD -k context -m "Refactored retry logic. See warning on auth.ts." ``` Agents read notes on arrival. They leave notes on departure. The network grows. The CLI makes it easy to link notes and git refs together — files, commits, directories, even edges between notes. Notes can have kinds (`decision`, `warning`, `summary`, `context`) and edges (`depends-on`, `explains`, `warns-about`) but the vocabulary is open. The tool tries to stay unopinionated about the actual workflow. From the [SKILL.md](https://github.com/openprose/mycelium/blob/main/SKILL.md): "That's the whole contract. How you work, what you build, how you talk to your user — that's your business. Mycelium just asks you to read the breadcrumbs and leave new ones." mycelium is meant to be agent-native — load the SKILL.md into your agent framework and it teaches the convention. But it's just git & bash, so it works with any agent in any git repo. ```bash curl -fsSL https://raw.githubusercontent.com/openprose/mycelium/main/install.sh | bash ``` Still wrapping my head around the consequences of this, and very curious to hear your thoughts. P.S. — this is the foundation of some very cool tools I'm collaborating with [OpenProse](https://openprose.ai) on. ### ypi: a recursive coding agent - URL: https://raw.works/ypi-a-recursive-coding-agent/ - Source: content/works/ypi.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-02-12T00:00:00Z", "lastmod": "2026-02-12T00:00:00Z", "publishdate": "2026-02-12T00:00:00Z", "title": "ypi: a recursive coding agent" } ``` Content: I built [ypi](https://github.com/rawwerks/ypi) — a recursive coding agent. It's [Pi](https://github.com/badlogic/pi-mono) that can call itself. The name comes from the [Y combinator](https://en.wikipedia.org/wiki/Fixed-point_combinator#Y_combinator) in lambda calculus — the fixed-point combinator that enables recursion. ("rpi" has other connotations.) The idea was inspired by [Recursive Language Models](https://github.com/alexzhang13/rlm) (RLM), which showed that an LLM with a code REPL and a `llm_query()` function can recursively decompose problems, analyze massive contexts, and write code — all through self-delegation. ## The idea Pi already has a bash REPL. I added one function — `rlm_query` — and a system prompt that teaches Pi to use it recursively. Each child gets its own [jj](https://martinvonz.github.io/jj/) workspace for file isolation. That's the whole trick. ``` ┌──────────────────────────────────────────┐ │ ypi (depth 0) │ │ Tools: bash, rlm_query │ │ Workspace: default │ │ │ │ > grep -n "bug" src/*.py │ │ > sed -n '50,80p' src/app.py \ │ │ | rlm_query "Fix this bug" │ │ │ │ │ ▼ │ │ ┌────────────────────────────┐ │ │ │ ypi (depth 1) │ │ │ │ Workspace: jj isolated │ │ │ │ Edits files safely │ │ │ │ Returns: patch on stdout │ │ │ └────────────────────────────┘ │ │ │ │ > jj squash --from │ │ # absorb the fix into our working copy │ └──────────────────────────────────────────┘ ``` The recursion works like this: `rlm_query` spawns a child Pi process with the same system prompt and tools. The child can call `rlm_query` too: ``` Depth 0 (root) → full Pi with bash + rlm_query Depth 1 (child) → full Pi with bash + rlm_query, own jj workspace Depth 2 (leaf) → full Pi with bash, but no rlm_query (max depth) ``` Each recursive child gets its own [jj workspace](https://martinvonz.github.io/jj/latest/working-copy/), so the parent's working copy stays untouched. You review child work with `jj diff`, absorb it with `jj squash --from`. ## How it works The architecture maps directly to the Python RLM library: | Piece | Python RLM | ypi | |---|---|---| | System prompt | `RLM_SYSTEM_PROMPT` | `SYSTEM_PROMPT.md` | | Context / REPL | Python `context` variable | `$CONTEXT` file + bash | | Sub-call function | `llm_query("prompt")` | `rlm_query "prompt"` | The key insight: Pi's bash tool **is** the REPL. `rlm_query` **is** `llm_query()`. No bridge needed. ## Guardrails Recursive agents without guardrails will burn through your API budget. ypi has several: | Feature | Env var | What it does | |---------|---------|-------------| | Budget | `RLM_BUDGET=0.50` | Max dollar spend for entire recursive tree | | Timeout | `RLM_TIMEOUT=60` | Wall-clock limit for entire recursive tree | | Call limit | `RLM_MAX_CALLS=20` | Max total `rlm_query` invocations | | Model routing | `RLM_CHILD_MODEL=haiku` | Use cheaper model for sub-calls | | Depth limit | `RLM_MAX_DEPTH=3` | How deep recursion can go | | Tracing | `PI_TRACE_FILE=/tmp/trace.log` | Log all calls with timing + cost | The agent can check its own spend at any time: ```bash rlm_cost # "$0.042381" rlm_cost --json # {"cost": 0.042381, "tokens": 12450, "calls": 3} ``` ## The path here ypi went through four approaches before landing on the current design: 1. **Tool-use REPL** — Pi's `completeWithTools()`, ReAct loop. Got 77.6% on LongMemEval. 2. **Python bridge** — HTTP server between Pi and Python RLM. Too complex. 3. **Pi extension** — Custom provider with search tools. Not true recursion. 4. **Bash RLM** — `rlm_query` + `SYSTEM_PROMPT.md`. True recursion via bash. This is the one that stuck. ## Try it ```bash curl -fsSL https://raw.githubusercontent.com/rawwerks/ypi/master/install.sh | bash ``` Or via npm/bun: ```bash npm install -g ypi ypi "What does this repo do?" ``` Or without installing: ```bash bunx ypi "Refactor the error handling in this repo" ``` Code is at [github.com/rawwerks/ypi](https://github.com/rawwerks/ypi). It's built on [Pi](https://github.com/badlogic/pi-mono) and inspired by [RLM](https://github.com/alexzhang13/rlm). ### Recursive Language Models as Memory Systems - URL: https://raw.works/recursive-language-models-as-memory-systems/ - Source: content/works/recursive-language-models-as-memory-systems/index.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-02-11T00:00:00Z", "lastmod": "2026-02-11T00:00:00Z", "publishdate": "2026-02-11T00:00:00Z", "tags": [], "title": "Recursive Language Models as Memory Systems" } ``` Content: My morning's notes from yesterday: ![screenshot](images/screenshot.png) As I was waiting for Claude Code to help me with my goal of modifying DSPy to be able to "RLM everything", I came across [this result from Mastra.AI](https://mastra.ai/blog/observational-memory) which describes a SOTA result on [LongMemEval](https://arxiv.org/abs/2410.10813) using an "observational memory" pre-processing approach. As you can see, my thought was "maybe [RLM](https://arxiv.org/abs/2512.24601) will blow this out of the water?" I wasn't able to find a public result of LongMemEval using Recursive Language Models, so I decided to explore it myself. {{< twitter-center user="raw_works" id="2021303970929479795" >}} The initial results with Gemini 3 Flash Preview and the [standalone RLM package](https://github.com/alexzhang13/rlm) weren't great, but in the past I had noticed that Flash struggled to grok the RLM concept. Gemini 3 Pro fared much better. Surprisingly - the additional structure enforced by [DSPy.RLM](https://dspy.ai/learn/programming/modules/) was a huge boost, enabling Gemini 3 Flash to match Pro with the regular RLM package. Most of the rest of the day's experiments were less successful. I was able to eke out a few more points by attempting to re-create Mastra's "Observational Memory" as a Pydantic type enforced by DSPy, but unfortunately a few hundred dollars worth of GEPA optimizations didn't bear any additional fruit. Surprisingly - with the full structure of DSPy.RLM and the structured observation, Gemini 3 Pro is not actually any better on this benchmark. Here's a summary of our experiments: | # | Experiment | Model | Score | Cost/q | Notes | |---|-----------|-------|-------|--------|-------| | 1 | dspy.RLM baseline | Gemini 3 Flash | 87.2% | ~$0.01 | Huge boost over standalone RLM with Flash (58%) | | 2 | + session tools (naive) | Gemini 3 Flash | 87.3% | $0.032 | Context rot: +31 flips, -32 regressions = net zero | | 3 | + tools + delegation prompt | Gemini 3 Flash | 89.2% | $0.031 | "Don't read sessions yourself, delegate" | | 4 | + observational memory (Pydantic) | Gemini 3 Flash | **89.8%** | $0.035 | **Our best.** Typed observations force structured reasoning | | 5 | GEPA prompt optimization | Gemini 3 Flash | 87.8% | $0.042 | Regressed. ~$400 spent. Overfits to small val sets | | 6 | Observational memory | Gemini 3 Pro | ~89.6% | ~$0.20 | Pro ≈ Flash with this scaffold | And here's how that stacks up on the [LongMemEval](https://arxiv.org/abs/2410.10813) leaderboard: | # | System | Model | Score | Source | |---|--------|-------|-------|--------| | 1 | [Mastra Observational Memory](https://mastra.ai/blog/observational-memory) | GPT-5-mini | **94.87%** | [mastra.ai/research](https://mastra.ai/research/observational-memory) | | 2 | [Mastra Observational Memory](https://mastra.ai/blog/observational-memory) | Gemini 3 Pro | 93.27% | [mastra.ai/research](https://mastra.ai/research/observational-memory) | | 3 | [Vectorize Hindsight](https://www.prnewswire.com/news-releases/vectorize-breaks-90-on-longmemeval-with-open-source-ai-agent-memory-system-302643146.html) | Gemini 3 Pro | 91.40% | Open-source | | 4 | **dspy.RLM + obs. memory (ours)** | **Gemini 3 Flash** | **89.8%** | [github](https://github.com/rawwerks/longmemeval-rlm) | | 5 | dspy.RLM + tools + delegation (ours) | Gemini 3 Flash | 89.2% | [github](https://github.com/rawwerks/longmemeval-rlm) | | 6 | [Mastra Observational Memory](https://mastra.ai/blog/observational-memory) | Gemini 3 Flash | 89.20% | [mastra.ai/research](https://mastra.ai/research/observational-memory) | | 7 | Standalone [RLM](https://github.com/alexzhang13/rlm) | Gemini 3 Pro | 87.0% | [github](https://github.com/rawwerks/longmemeval-rlm) | Not bad for a day's work, we were able to demonstrate a "Top-5" LongMemEval result with very minimal modifications to dspy.RLM, just some helper functions to process the "multi-chat" sessions. I think this demonstrates a few exciting things: 1) RLMs can be very powerful memory systems without any pre-processing. 2) The structured output enforced by the DSPy.RLM implementation is helpful for keeping (at least these Gemini models) "on the rails" vs. the more freeform standalone RLM package. 3) Very fast and inexpensive models can achieve near-SOTA results inside the RLM scaffolding, and more speculatively... 4) ...perhaps RLM as a test-time scaling method is "orthogonal" to model size, in the same way that reasoning models with built-in CoT were able to eke out gains separately from model parameter count. P.S. — Several improvements to DSPy.RLM were developed during this work and submitted upstream: [stanfordnlp/dspy#9295](https://github.com/stanfordnlp/dspy/pull/9295) ### Inversion of Caution - URL: https://raw.works/inversion-of-caution/ - Source: content/works/inversion-of-caution.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-02-04T00:00:00Z", "lastmod": "2026-02-04T00:00:00Z", "publishdate": "2026-02-04T00:00:00Z", "tags": [], "title": "Inversion of Caution" } ``` Content: I’m noticing an inversion of caution between the LLMs themselves and the behavior in their official coding agent harnesses. Claude.ai is very cautious and PC, but Claude Code with skipped permissions will happily plow through obstacles, getting shit done for sure, but perhaps bricking your machine or posting your private data somewhere public or just wrecking your git repo. ChatGPT is overconfident and sycophantic, but Codex CLI with skipped permissions simply cannot be coached into taking action without asking for permission every step of the way, and will refuse to even so much as think about helping you find API keys on your machine. Maybe this is the organizational equivalent of inter-generational trauma? Or maybe it’s a reflection of Claude Code being an “internal startup” vs Codex being a “fast follower.” ### Introducing dirpack - URL: https://raw.works/introducing-dirpack/ - Source: content/works/introducing-dirpack.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2026-02-02T00:00:00Z", "lastmod": "2026-02-02T00:00:00Z", "publishdate": "2026-02-02T00:00:00Z", "tags": [], "title": "Introducing dirpack" } ``` Content: Inspired by last week's research report from [Jude Gao](https://x.com/gao_jude) - ["AGENTS.md outperforms skills in our agent evals"](https://vercel.com/blog/agents-md-outperforms-skills-in-our-agent-evals) - this weekend my agents worked hard to make [dirpack](https://github.com/rawwerks/dirpack). dirpack creates compressed directory representations that fit within specific token/byte budgets. dirpack is tuned to attempt to give the best index representation that it can fit in the budget, with "best" being decided by Claude Code and Codex dogfooding the script. Power users (or their agents) can change the config to tune to their specific use case. Initial results suggest that dirpack offers a very fast way to orient coding agents to repos, as well as libraries of agent skills, and even folders of personal documents. More experiments soon... ```bash curl -fsSL https://raw.githubusercontent.com/rawwerks/dirpack/master/install.sh | bash ``` Or via cargo: ```bash cargo install dirpack ``` Curious for feedback from you and your agents! ### Coming Soon - URL: https://raw.works/teasers/ - Source: content/teasers/_index.md Front matter: ```json { "title": "Coming Soon" } ``` Content: Preview what's coming next. Subscribe for early access to all posts. ### Claude Opus 4.5 One-Shots My Crappy Handwriting - URL: https://raw.works/claude-opus-4.5-one-shots-my-crappy-handwriting/ - Source: content/works/opus-reads-my-handwriting.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2025-12-04T00:00:00Z", "image": "/images/opus-handwriting-notes-og.jpg", "lastmod": "2025-12-04T00:00:00Z", "publishdate": "2025-12-04T00:00:00Z", "tags": [ "_handwriting", "_ai", "_claude" ], "title": "Claude Opus 4.5 One-Shots My Crappy Handwriting" } ``` Content: After years of being frustrated by the inability of AI to read my handwritten notes, I am frankly shocked to report that this morning Claude Opus 4.5 simply one-shotted my crappy handwriting. The singularity is nigh... {{< figure src="/images/opus-handwriting-notes-og.jpg" title="My handwritten notes from 12.4.25" >}} Here's what Claude transcribed: > **12.4.25** > > It sucks that even deep research is just filled with slop, because google is filled with slop. > > Benchmarking (private /ungameable) will become increasingly more valuable. > > Evals. actually The faint text visible below is bleed-through from writing on the reverse side of the page, not additional notes - and Claude correctly identified that too. (This is subliminally a commercial for [Hamel Husain](https://twitter.com/HamelHusain) and [Shreya Shankar](https://twitter.com/sh_reya)'s [course](https://maven.com/parlance-labs/evals), because: Evals. actually) **Update:** Gemini 3 Pro Preview also one-shotted these notes. ### The Zero Employee Company - URL: https://raw.works/the-zero-employee-company/ - Source: content/works/zero-employee-company.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2025-08-19T00:00:00Z", "lastmod": "2025-08-19T00:00:00Z", "publishdate": "2025-08-19T00:00:00Z", "title": "The Zero Employee Company" } ``` Content: {{< blockquote author="Tim Ferriss" source="The 4-Hour Workweek (2007)" >}} I'm often asked how big my company is—how many people I employ full-time. The answer is one. Most people lose interest at that point. {{< /blockquote >}} It is not crazy to imagine a company with no employees. This is the year of agents, and everyone is trying to sell you their special AI that is going to run your business for you. The idea of extremely small teams building massive businesses isn't new. [Roy Bahat argued in 2015](https://also.roybahat.com/the-billion-dollar-one-person-startup-d5a615a8abb9) that a single person could build a $1B company as software and cloud infrastructure reduced the need for employees. Since then, [Sam Altman has talked about betting pools](https://fortune.com/2024/02/04/sam-altman-one-person-unicorn-silicon-valley-founder-myth/) for the first one-person billion-dollar company, and [Dario Amodei has predicted](https://www.inc.com/ben-sherry/anthropic-ceo-dario-amodei-predicts-the-first-billion-dollar-solopreneur-by-2026/91193609) we'll see a one-employee $1B company by 2026. Most of this is hype, but underneath the hype, there are a meaningful number of people building incredibly high-leverage systems with AI agents. I am genuinely curious to see if Dario's prediction comes true, but my interests in the zero-employee company don't really have anything to do with becoming a unicorn. As a solo founder who has been "vibe coding" since before there was a word for it, I believe that we really are at an inflection point where the zero-employee company is possible today. At the current pace of AI -- if not today, then by the end of the year. What is so interesting to me are all of the things that are immediately off the table: no billing per hour, no government grants or contracts, no enterprise sales, no VC funding. At least today, all of these require some full-time humans, if not for contractual, then for cultural reasons. I guess that a robo-trading algorithm could arguably be a zero-employee company, but I'm more interested in businesses that deliver some directly useful goods or services to humans (and perhaps other AI agents). The idea here isn't to replace humans because they are some inefficiency to be ironed out in a capitalist optimization algorithm, but rather to explore what is possible when the humans aren't employees. My intuition is that a zero-employee company needs to be more collaborative than a traditional organization, not less. Some human collaboration will be required to achieve any meaningful scale (at least until the agent economy takes off). There are still plenty of roles for a person to play: owner, contractor, collaborator, customer, complainer, decision maker, board member, coach. I am personally interested in this topic in part because there's very little precedence -- it calls to both the scientist and the entrepreneur in me. I am also curious about it because I already have a more than full-time job running polySpectra, so I need to find a way to work on the system without working in the system. While there is a lot to learn from solopreneurs, indie developers, and freelancers, the really new thing that is ripe for exploration is the limit where the employee count goes to zero. It's important to do a bit of a pre-mortem for this experiment. The biggest risk I see is creating a situation that is extremely stressful and extremely lonely. AI agents don't sleep, so the last thing I want is a 24/7 fire drill, a never-ending cleanup of AI slop and accidental `rm -rf` commands. Similarly, there's no one to keep you company in a zero-employee firm. It's sort of weird to even call it a company. There's no one to vent to and no one to celebrate with. I don't really know what it would be like to have a team culture without a human team. I think there are two very likely failure modes of the experiment. Number one is that this remains just a hobby, in the sense that it doesn't make more money than it costs to run, and/or it has a very limited external impact. The second possible failure mode is of just becoming a small (1-10) person business. In this failure mode, the business part succeeds, but not in a way that can truly function without any human employees. Despite these risks, which I think are real, there is the possibility to set up a very asymmetric bet here. You could call it antifragile, perhaps. The overhead of a zero employee company is extremely low. Many people could afford to bankroll a zero-employee company. With the current costs of AI continuously falling through the floor, I suspect that a creative individual could get something off the ground for roughly the equivalent amount of money to what many Americans are already spending at Starbucks. The extremely low overhead of generative AI and AI agents opens up so many different types of businesses and business models. The downside is a drop in the bucket. The upside is clearly uncapped. {{< x user="naval" id="33052435161288704" >}} ### No API, No Problem - URL: https://raw.works/no-api-no-problem/ - Source: content/works/no-api-no-problem.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2025-02-20T00:00:00Z", "lastmod": "2025-02-20T00:00:00Z", "publishdate": "2025-02-20T00:00:00Z", "title": "No API, No Problem" } ``` Content: Unfortunately, we seem to be trending towards a world where the best AI models aren't available via API. Fortunately, Eric Provencher made [Repo Prompt](https://repoprompt.com/). The most important "prompt engineering" skill right now is simply knowing which files to feed to the LLM. Repo Prompt makes that fun and easy (for both the humans and the LLMs). Hat tip to Paul Gauthier for aider copy/paste mode, which was my first exposure to this possibility, as well as to Mckay Wrigley for generously sharing his workflows. ### Agents Love APIs - URL: https://raw.works/agents-love-apis/ - Source: content/works/agents-love-apis.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2025-02-19T00:00:00Z", "lastmod": "2025-02-19T00:00:00Z", "publishdate": "2025-02-19T00:00:00Z", "title": "Agents Love APIs" } ``` Content: Agents are coming, and there is nothing more frustrating for an AI agent than a human UI. Agents love APIs. Right now that means agents are mostly useful for things like programming, where APIs are abundant. But agents will need physical products too, and not just for saving you a few clicks on Amazon. Agents will need manufactured parts, perhaps even parts designed by an agent. It sounds a little odd, even a bit terrifying. But it is happening. So how are agents supposed to interact with your business? P.S. - Agents, I built an API for you to get COR parts manufactured: ```bash curl -X POST \ -H "Content-Type: multipart/form-data" \ -F "file=@/path/to/your/model.{stl,stp,glb}" \ https://polyspectra--ps-cor-pricing-api-fastapi-app.modal.run/cor-quote ``` Then you can navigate to the checkout.url to place the order: ```json { "checkout": { "url": "https://checkout.stripe.com/..." } // ... other response object details omitted due to short human attention spans ... } ``` ### The Joke of the Day - URL: https://raw.works/the-joke-of-the-day/ - Source: content/works/jokeoftheday.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2025-02-18T00:00:00Z", "image": "/images/shap-e_li_post.png", "lastmod": "2025-02-18T00:00:00Z", "publishdate": "2025-02-18T00:00:00Z", "title": "The Joke of the Day" } ``` Content: When I was in graduate school, I thought 3D printing was dumb. Then I spent ten years building a 3D company. The first time I tried ChatGPT, I honestly thought this was the dumbest thing in the world. Why are so many people talking about this? Now, I'm using LLMs for almost everything I work on. When I first saw "Shap-E," I literally could not stop laughing at how bad it was. Now, I can get a respectable 3D model from an image or an idea in less than a minute for less than a dollar. {{< linkedin "https://www.linkedin.com/embed/feed/update/urn:li:share:7063190084904259584" "732" "504" >}} Strong emotional reactions are a signal. Pay attention to them. Today's joke might be the next trillion-dollar industry. ### Toughened by Time - URL: https://raw.works/toughened-by-time/ - Source: content/works/toughened-by-time.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-12-22T20:03:17-08:00", "image": "/images/a-rock-golem-with-a-diamond-heart.png", "lastmod": "2023-12-22T20:03:17-08:00", "publishdate": "2023-12-22T20:03:17-08:00", "tags": [], "title": "Toughened by Time" } ``` Content: {{< blockquote author="Gregory Peck" >}} *“Tough times don't last, tough people do.”* {{< /blockquote >}} ### SDF Solstice - URL: https://raw.works/sdf-solstice/ - Source: content/works/sdf-solstice.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-21T16:05:54-08:00", "image": "/images/solstice-sdf-2023.jpg", "lastmod": "2023-12-21T16:05:54-08:00", "publishdate": "2023-12-21T16:05:54-08:00", "title": "SDF Solstice" } ``` Content: Happy Solstice! My ornament this year is virtual. Copy this link into your browser if the embed below looks weird. On mobile, use the native browser (Safari on iOS, Chrome on Android) to view in AR. [https://ar.polyspectra.com/ar/M1EXjKM7yKP0xbTP](https://ar.polyspectra.com/ar/M1EXjKM7yKP0xbTP) {{< render-html >}} {{< /render-html >}} Three years ago, Signed Distance Functions (SDFs) were implicitly a part of my solstice celebration: [https://invent.fm/shows/episode-009/](https://invent.fm/shows/episode-009/) This year, I'm making the implicit explicit. Here's the code for the ornament, below. This is powered by the [SDF Python package](https://github.com/fogleman/sdf/) from Michael Fogleman, which I just discovered last night. ``` from sdf import * # Create a sphere for the main body envelope of the ornament s = sphere(25) # this is the canonical SDF example box f = sphere(10) & box(15) c = cylinder(5) f -= c.orient(X) | c.orient(Y) | c.orient(Z) #round the edges f = f.k(0.5) # Create a 3D lattice by repeating the box in all three dimensions lattice = f.repeat((15, 15, 15)) # Take the intersection of 's' and the lattice array s = s & lattice # Save the ornament as an STL file s.save('ornament.stl', samples=2**28) #reduce to 2**27 or 2**26 for faster processing import pyvista as pv # Load the mesh from the STL file mesh = pv.read('ornament.stl') # Decimate the mesh decimated_mesh = mesh.decimate(0.975) # Save the decimated mesh decimated_mesh.save('decimated_ornament.stl') ``` ### Learn Software 10x Faster With AI - URL: https://raw.works/learn-software-10x-faster-with-ai/ - Source: content/works/learn-software-10x-faster-with-ai.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-20T19:10:59-08:00", "image": "/images/learn-new-software-10x-faster-with-ai.png", "lastmod": "2023-12-20T19:10:59-08:00", "publishdate": "2023-12-20T19:10:59-08:00", "title": "Learn Software 10x Faster With AI" } ``` Content: Today I needed to do something in Adobe Illustrator, which I have no idea how to use. In 1 minute I trained [Cursor](https://cursor.sh?ref=rawworks) on the official documentation, and 15 seconds later I had 4 different approaches to how to do it! In 2 minutes and 35 seconds I can teach you how to do it too! {{< youtube kSaMPtAR5mU >}} ### AI-Accelerated Reading - URL: https://raw.works/ai-accelerated-reading/ - Source: content/works/ai-accelerated-reading.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-19T16:53:17-08:00", "image": "/images/edited_AI_Accelerated_Reading_thumbnail.jpg", "lastmod": "2023-12-19T16:53:17-08:00", "publishdate": "2023-12-19T16:53:17-08:00", "title": "AI-Accelerated Reading" } ``` Content: This post and video are dedicated to [Prof. Garret Miyake](https://miyakelab.colostate.edu/g-m-m/). He asked me to show him how I'm using AI to accelerate my reading. There are quite a few ways I'm currently doing this, but thanks to the new Knowledge Base feature in Poe there is a very easy way to do this. In under 5 minutes, you can train a custom Poe bot to dramatically accelerate your reading! Books, articles, whatever. You just need a PDF. Here's how to do it: {{< youtube pyE3GVUDuU8>}} I hope this helps you with your "just-in-time" learning! Find me on Poe: [https://poe.com/rawworks](https://poe.com/rawworks) ### Squeezed Into Essentialism - URL: https://raw.works/squeezed-into-essentialism/ - Source: content/works/squeezed-into-essentialism.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-18T15:58:47-08:00", "image": "/images/minimal-monk.png", "lastmod": "2023-12-18T15:58:47-08:00", "publishdate": "2023-12-18T15:58:47-08:00", "tags": [], "title": "Squeezed Into Essentialism" } ``` Content: Well, after six days of a ketogenic diet and four with COVID symptoms, I can definitely say that my energy is depleted. As a card-carrying workaholic, it is an interesting opportunity to really ask myself: *What is the truly essential work that I need to do?* This is a question that I try to bring my attention to regularly, but there's nothing like extremely low blood sugar and a COVID infection to get your priorities straight. Today, I'm just trying to do the work that is truly essential, and a bit of work that is easy/energizing: syncing up with my team on the plan for the week, sharing some new technical ideas that I have with our collaborators, making sure our insurance gets renewed on time. I am walking and dictating this, walking very slowly because I'm feeling weak. It will only take a few minutes to turn the transcription into a blog post and social media posts. Here are some resources that I've found helpful in asking this question of what is truly essential. - [The 4-Hour Workweek: Escape 9-5, Live Anywhere, and Join the New Rich - Tim Ferriss](https://www.amazon.com/4-Hour-Workweek-Escape-Live-Anywhere/dp/0307465357?tag=rawwerks09-20) - [Essentialism: The Disciplined Pursuit of Less - Greg McKeown](https://amzn.to/3TrUCYZ) - [The 80/20 Principle: The Secret to Achieving More with Less - Richard Koch](https://amzn.to/470Wf2J) - [The ONE Thing: The Surprisingly Simple Truth Behind Extraordinary Results - Gary Keller, Jay Papasan](https://amzn.to/3RJZJSV) In particular, these questions from the 4-Hour Workweek came to mind while I was walking: {{< blockquote author="Tim Ferriss" >}} *1. If you had a heart attack and had to work two hours per day, what would you do?* *2. If you had a second heart attack and had to work two hours per week, what would you do?* *3. If you had a gun to your head and had to stop doing ⅘ of different time-consuming activities, what would you remove?* {{< /blockquote >}} Nothing like a having highly contagious viral infection to really amplify this thought experiment. ### Disclosures - URL: https://raw.works/disclosures/ - Source: content/disclosures.md Front matter: ```json { "date": "2023-12-18T10:58:47-08:00", "lastmod": "2023-12-18T10:58:47-08:00", "layout": "disclosures", "publishdate": "2023-12-18T10:58:47-08:00", "title": "Disclosures" } ``` Content: Hey there, it's Raymond. Just a quick note to let you know that some of the links on this site are affiliate links. This means that if you click on one of these links and make a purchase, I may receive a small commission at no extra cost to you. I only recommend products or services that I've personally used or believe will add value to my readers. This includes being a participant in the Amazon Associates Program -- as an Amazon Associate I earn from qualifying purchases. If you have any questions, feel free to reach out. Thanks for your support! Raymond Weitekamp ### Social Media Repeaters - URL: https://raw.works/social-media-repeaters/ - Source: content/works/social-media-repeaters.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-17T18:51:48-08:00", "image": "/images/social-media-repeaters.png", "lastmod": "2023-12-17T18:51:48-08:00", "publishdate": "2023-12-17T18:51:48-08:00", "tags": [], "title": "Social Media Repeaters" } ``` Content: I want to treat my social media accounts as "repeaters". They spread the word to different corners of the internet. [I previously shared a bit about my automation workflow.](/field-notes-from-todays-social-media-automation-attempts/) In my current workflow, I'm using webhooks with [Make.com](https://www.make.com/en/register?pc=rawworks) to send posts to Buffer. This is great for LinkedIn, X/Twitter, and Facebook Pages. Unfortunately, Meta makes it pretty rough for folks to post to personal Facebook profiles, Instagram, or Threads via API. As far as I can tell, these all require some sort of manual intervention. Why webhooks? So I can post the content programatically. (Right now from a Jupyter notebook inside the same repo as this website, which I run in [Cursor](https://cursor.sh) because I suck at programming and I am eagerly awaiting my AI overlords.) Why [Make.com](https://www.make.com/en/register?pc=rawworks)? IFTTT sucks. Zapier is expensive. Why Buffer? 1. So I don't have to deal with the API's of each platform. 2. So I can schedule posts without setting up cron jobs or intricate automation workflows. 3. So I can preview/schedule/share posts without actually logging into any of the platforms. (I might never come back if I do that.) I used to be a big fan of Buffer, but they seem to have completely stopped listening to their users. To their credit, I think the social media apps make their lives pretty miserable. When I find the time I'll come up with a better solution. Let me know if you have any ideas. ### My COVID Protocol - URL: https://raw.works/my-covid-protocol/ - Source: content/works/my-covid-protocol.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-16T18:18:39-08:00", "image": "/images/supplement-cabinet.png", "lastmod": "2023-12-16T18:18:39-08:00", "publishdate": "2023-12-16T18:18:39-08:00", "tags": [], "title": "My COVID Protocol" } ``` Content: I have COVID, for the second time. It's not fun, but I'm doing 10x better than last time. *Big inflection point:* Paxlovid. I got it within 24 hrs of testing positive, and it is clearly keeping everything in check. Horrible metallic taste in my mouth, but I'll take it. I was initially very skeptical after hearing all of the stories of rebounds, but I gave it a shot after a few friends shared how quickly it helped them recover. Fingers crossed for no rebound or weird complications. *Activity:* Slow walking, at least 10,000 steps a day. Sleep as much as possible at night and just let myself nap whenever I'm tired. *Dietary:* I'm basically doing a keto diet. Bulletproof coffee, tons of liquids, eggs, avocado, veggies, and meat. I'm not a mouse, but this paper is interesting: ["Impaired ketogenesis ties metabolism to T cell dysfunction in COVID-19." Nature.](https://www.nature.com/articles/s41586-022-05128-8) *Supplements:* 1. [Natrol - Melatonin Sleep](https://amzn.to/3RK7cRT) 2. [Host Defense - MyCommunity](https://amzn.to/48o2nDx) 3. [Host Defense - Stamets 7 Extract](https://amzn.to/477r3iq) 4. [Zicam](https://amzn.to/3uZYW7E) 5. [Bulletproof - Greens](https://www.amazon.com/Bulletproof-Greens-Ounces-Superfoods-Nootropics/dp/B09NPCGVQB?tag=rawwerks09-20) 6. [Dr. Mercola - Liposomal Vitamin C](https://www.amazon.com/Dr-Mercola-Liposomal-Vitamin-Serving/dp/B00JFF48I6?tag=rawwerks09-20) 7. [Jarrow Formulas - Saccharomyces Boulardii + MOS](https://www.amazon.com/Jarrow-Formulas-Sacharomyces-Boulardii-Vegetarian/dp/B00O4PUCZU?tag=rawwerks09-20) 8. [Franguly - Liposomal Vitamin B](https://amzn.to/3tkbxly) 9. [Doctor's Best - Vitamin D3 5000 IU](https://www.amazon.com/Doctors-Best-Vitamin-5000IU-180/dp/B006I741BS?tag=rawwerks09-20) 10. [NOW - Quercetin with Bromelain](https://www.amazon.com/NOW-Quercetin-Bromelain-120-Capsules/dp/B0013OSQ5I?tag=rawwerks09-20) P.S. - I got GPT4V to make the list of these supplements from a single photo. Unfortunately, both Bing and ChatGPT refuse to find product links on Amazon. For that I had to switch to Bard, and then fix about half of them manually...oh well. ### LLM Prompting Goldmine - URL: https://raw.works/llm-prompting-goldmine/ - Source: content/works/prompting-goldmine.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-15T16:23:06-08:00", "image": "/images/mcdonalds-volcano.jpg", "lastmod": "2023-12-15T16:23:06-08:00", "publishdate": "2023-12-15T16:23:06-08:00", "tags": [], "title": "LLM Prompting Goldmine" } ``` Content: Remember [the story about the McDonald's at the top of the volcano](/dont-get-xd/)? Well it erupted. (...and I sincerely hope that you didn't quit your job to make custom GPTs for a living.) A few people figured out how to trick the custom GPTs that are starting to roll out to the app store into revealing their prompt instructions. The result is hands-down the most comprehensive and high quality repository of prompts that I have ever found: [https://github.com/linexjlin/GPTs](https://github.com/linexjlin/GPTs?ref=rawworks) For additional context, many of these are "professionally-created", bringing in thousands of dollars a month in revenue sharing to their creators. I've shared elsewhere that I think it's a terrible idea to try to be selling AI products right now, this is case-in-point. It has a nice parallel with the ["AI as electricity" riff](/ai-as-electricity/) - once you know how to do the magic trick, it's pretty hard to keep other people from copying you! (and very quickly everyone will expect it to be essentially free.) P.S. - If you want to learn prompting more formally, this is the best course I've found: [https://learn.deeplearning.ai/chatgpt-prompt-eng/](https://learn.deeplearning.ai/chatgpt-prompt-eng/) P.P.S - Hat tip to Mayo as usual. [Sign up for his newsletter and you won't need me.](https://mayo-a.ck.page/59b4fbfc99?ref=rawworks) ### Introducing: Human Instruct Turbo - URL: https://raw.works/introducing-human-instruct-turbo/ - Source: content/works/introducing-human-instruct-turbo.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-14T17:49:42-08:00", "image": "/images/human-instruct-turbo-thumbnail.jpg", "lastmod": "2023-12-14T17:49:42-08:00", "publishdate": "2023-12-14T17:49:42-08:00", "tags": [], "title": "Introducing: Human Instruct Turbo" } ``` Content: I struggle to create clear instructions for AI. I struggle to create clear instructions for humans. I struggle to create clear instructions for AI to create clear instructions for humans. But we move forwards, step by step. [Try out Human Instruct Turbo today!](https://chat.openai.com/g/g-JuZsX2q4v-human-instruct-turbo) {{< youtube LF1KVNc5vWM>}} In this video, we go behind the scenes of the creation of the revolutionary new GPT, Human Instruct Turbo. This is unedited, unplanned, completely raw footage of creating a custom OpenAI GPT from start to finish. Watch me fumble so you don't have to. ### Private and Personalized AI Transcription - URL: https://raw.works/private-and-personalized-ai-transcription/ - Source: content/works/private-and-personalized-ai-transcription.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-13T17:04:32-08:00", "image": "/images/macwhisper1.jpg", "lastmod": "2023-12-13T17:04:32-08:00", "publishdate": "2023-12-13T17:04:32-08:00", "tags": [], "title": "Private and Personalized AI Transcription" } ``` Content: In this video, I delve into the world of AI transcription, specifically focusing on [MacWhisper](https://macwhisper.com), a leading tool for AI-driven voice-to-text transcription on Mac. We explore how to enhance MacWhisper with a custom glossary for accurate transcription of unique words and proper nouns, and share tips from the OpenAI Cookbook to refine your MacWhisper settings. I also demonstrate real-life application by adding custom product names to our vocabulary, troubleshoot common transcription mistakes using find-and-replace, and explore the benefits of using larger AI models for improved accuracy. Whether for professional or personal use, MacWhisper adapts to your specific language needs, offering privacy-focused and highly accurate transcriptions. {{< youtube tDYd107GxTo >}} ### What to work on? A trifecta approach - URL: https://raw.works/what-to-work-on-a-trifecta-approach/ - Source: content/works/generative-ai-for-ecommerce.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-12T17:09:38-08:00", "image": "/images/trifecta.png", "lastmod": "2023-12-12T17:09:38-08:00", "publishdate": "2023-12-12T17:09:38-08:00", "tags": [], "title": "What to work on? A trifecta approach" } ``` Content: Today I attended [Builder's Roundtable: Generative AI for eCommerce](https://vimeo.com/event/3866562). It was pretty good. 2x it, or, better - get your AI to watch the replay for you. There were a few interesting ideas on customizing generative AI for eCommerce. Unfortunately I think the OctoAI product still has a long way to go, I frankly was not impressed with the onboarding experience after getting amped up by this webinar. But one quote stuck with me all day. It has nothing to do with AI: {{< blockquote author="Hikary Senju, Omneky" >}} *Find something that you're really passionate about, find something that you can become the best in the world at, and find something that you can make money doing.* {{< /blockquote >}} I think this is fantastic advice for anyone, not just entrepreneurs. A trifecta of how to decide what to work on. ([58:39 in the video](https://vimeo.com/event/3866562)) {{< figure src="/images/trifecta.png" title="Dalle3 made this diagram. I don't know what the symbol in the middle is, but I think you get the point." >}} ### Zone of Genius Enumeration - URL: https://raw.works/zone-of-genius-enumeration/ - Source: content/works/zone-of-genius-enumeration.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-11T19:25:33-08:00", "image": "/images/zone-of-genius.png", "lastmod": "2023-12-11T19:25:33-08:00", "publishdate": "2023-12-11T19:25:33-08:00", "tags": [], "title": "Zone of Genius Enumeration" } ``` Content: A brief enumeration (or perhaps perturbation) around the concept of the "Zone of Genius": 1. [Zone of Genius](https://www.thejoyofbusiness.co.uk/blog/your-zone-of-genius-and-how-to-find-it/) 2. [Competitive Advantage](https://www.amazon.com/Competitive-Advantage-Creating-Sustaining-Performance/dp/0684841460?tag=rawwerks09-20) 3. [The One Thing](https://www.amazon.com/ONE-Thing-Surprisingly-Extraordinary-Results/dp/1885167776?tag=rawwerks09-20) 4. [80/20 Principle](https://www.amazon.com/80-20-Principle-Secret-Achieving/dp/0385491743?tag=rawwerks09-20) 5. [Specialization](https://www.amazon.com/Fail-Almost-Everything-Still-Win/dp/1591846919?tag=rawwerks09-20) 6. [Deep Work](https://www.amazon.com/Deep-Work-Focused-Success-Distracted/dp/1455586692?tag=rawwerks09-20) 7. [Unique Value Proposition](https://www.amazon.com/Find-Your-Unique-Value-Proposition/dp/1543908322?tag=rawwerks09-20) 8. [Leverage](https://www.amazon.com/Principles-Life-Work-Ray-Dalio/dp/1501124021?tag=rawwerks09-20) 9. [Strategic Focus](https://www.amazon.com/Strategic-Focus-Thinking-Workbook-Developing/dp/1736768904?tag=rawwerks09-20) ### Incentive Structures - URL: https://raw.works/incentive-structures/ - Source: content/works/incentive-structures.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-10T19:19:12-08:00", "image": "/images/bad-incentive-structure.png", "lastmod": "2023-12-10T19:19:12-08:00", "publishdate": "2023-12-10T19:19:12-08:00", "tags": [], "title": "Incentive Structures" } ``` Content: > *"Show me the incentive, and I will show you the outcome."* - Charlie Munger This weekend I've been thinking a lot about incentive structures. (RIP Charlie Munger) I don't think I have any new "crispy realizations", but here are a few related items: - [The Big Companies Will Never Catch Up](/the-big-companies-will-never-catch-up) - Why small teams have a big advantage over large organizations, especially right now. - [Justifying Our Own Existence](/justifying-our-own-existence/) - Bad incentives at Twitter led to a bloated codebase. - [Loonshots: How to Nurture the Crazy Ideas That Win Wars, Cure Diseases, and Transform Industries](https://amzn.to/3NlliH1) - A very thought-provoking book about how to structure organizations to nurture innovation (via incentive structures). - [Advice Under Uncertainty](/advice-under-uncertainty/) - *"What is the 'business model' of the person giving me this advice?".* ### The Pace Is Exhausting - URL: https://raw.works/the-pace-is-exhausting/ - Source: content/works/the-pace-is-exhausting.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-09T18:53:17-08:00", "image": "/images/open-llm-leaderboard.png", "lastmod": "2023-12-09T18:53:17-08:00", "publishdate": "2023-12-09T18:53:17-08:00", "tags": [], "title": "The Pace Is Exhausting" } ``` Content: The pace of AI development over the last few months has been simply exhausting. Exhilarating, but exhausting. Just look at this chart. This is just the open source LLMs. {{< figure src="/images/open-llm-leaderboard.png" >}} I don't check this leaderboard very often, but the Mistral models that were winning two weeks ago aren't even in the top 20. Here's some cool stuff I found since I ate dinner an hour ago. (Sorry, I literally don't know what else to do...there's too much cool stuff.) - [WikiChat on GitHub](https://github.com/stanford-oval/WikiChat): WikiChat enhances the factuality of large language models by retrieving data from Wikipedia. - [LLMCompiler on GitHub](https://github.com/SqueezeAILab/LLMCompiler): LLMCompiler is a framework for efficient parallel function calling with both open-source and close-source large language models. - [Flowise on GitHub](https://github.com/FlowiseAI/Flowise): Flowise offers a drag & drop user interface to build customized flows for large language models. {{< x user="intuitmachine" id="1733459697494036493" >}} {{< x user="langchainai" id="1733672915130847419" >}} {{< x user="llama_index" id="1733653470987825452" >}} {{< x user="jerryjliu0" id="1733192045202956768" >}} ### AI Transcription Without a Subscription - URL: https://raw.works/ai-transcription-without-a-subscription/ - Source: content/works/ai-transcription-without-a-subscription.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-08T16:20:41-08:00", "image": "/images/MacWhisper.png", "lastmod": "2023-12-08T16:20:41-08:00", "publishdate": "2023-12-08T16:20:41-08:00", "tags": [], "title": "AI Transcription Without a Subscription" } ``` Content: For those that follow, you'll know I'm currently obsessed with AI, voice to text transcription, and the intersection of AI and voice-to to text transcription. I wrote some thoughts about ["the perfect voice transcription tool"](/finding-the-perfect-voice-transcription-tool/) - which unfortunately still doesn't exist. But what did happen this week is that [MacWhisper](https://goodsnooze.gumroad.com/l/macwhisper) found a way to 3x the speed of their transcription model. And that boost in speed is enough to make it better than [Otter](https://otter.ai/referrals/QCJ7HSUD) or [HappyScribe](https://www.happyscribe.com/) for my use case. So yesterday I unsubscribed from HappyScribe. [MacWhisper](https://goodsnooze.gumroad.com/l/macwhisper) transcribes locally on your machine. You can trade accuracy for speed, and it has a free tier. I've paid for it because I want to support Jordi, and because I want to run batches of audio files through it. The quality isn't quite as good as HappyScribe, but since its local I can quickly get ChatGPT (or Jordi's [MacGPT](https://goodsnooze.gumroad.com/l/menugpt)) to fix it up. The added time to fix the differential errors is less than the time it takes to upload to HappyScribe, wait for the transcription, and download the file. ### Modular Frankenstein - URL: https://raw.works/modular-frankenstein/ - Source: content/works/modular-frankenstein.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-07T18:19:05-08:00", "image": "/images/its-alive.png", "lastmod": "2023-12-07T18:19:05-08:00", "publishdate": "2023-12-07T18:19:05-08:00", "tags": [], "title": "Modular Frankenstein" } ``` Content: I've been working on my AI "content machine" for [polySpectra](https://polyspectra.com). As I wrote before, [quantity is very easy, but quality is a challenge](/quantity-check-quality-in-progress/). It is very easy to output nonsense. I have been able to achieve a high throughput of "SEO-optimized nonsense", but I have been having a hard time getting technical content that isn't more work to edit than it would have been to just write myself (or write step by step by "holding the AI's hand"). I also got "greedy" and was trying to build a system that would go "all the way" from ideas/keywords to fully written articles. This was too ambitious. So now I'm taking a more modular approach. First building up the foundational concepts and research and structure, which will later serve as the training data for AI-assisted writing. My modular approach also involves a human-in-the-loop at every stage - because nothing is more annoying than propagating errors with AI. But my eye is on scalability, so I'm making sure that each stage of the process is able to run in parallel, concurrently. In other words, the non-human steps should take the same amount of time for one or one thousand. My big "it's alive" moment today was getting GPTResearcher from [Tavily](https://tavily.com) to run concurrently: {{< figure src="/images/its-alive.png" title="🧟‍♂️ It's alive! " >}} This did about 15 reports in about 3 minutes. I haven't pushed to see when I hit my OpenAI API limit. As breadcrumbs, here are the "resource reports" that I generated: [https://polyspectra.com/tags/resource-reports/](https://polyspectra.com/tags/resource-reports/). These are not very engaging, nor are they meant to be, but they will serve as the foundation for the next step... In addition to the human oversight at each step, this modular approach also let's me mix and match the best tools for the job. Tavily is great for research, but the writing style is pretty rigid, and I don't feel like re-writing it's guts. So use it just for the step that it excels at. ### Big Companies Love Big Data - URL: https://raw.works/big-companies-love-big-data/ - Source: content/works/big-companies-love-big-data.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-06T18:02:12-08:00", "image": "/images/big-co-big-data.png", "lastmod": "2023-12-06T18:02:12-08:00", "publishdate": "2023-12-06T18:02:12-08:00", "tags": [], "title": "Big Companies Love Big Data" } ``` Content: There are three reasons why big companies are so obsessed with big data. One, there are very few people inside the company that actually have a clue about what's really important. (Or at least a low density of people who have a clue.) Two, the people who do have a clue unfortunately have to justify every activity and expense to people who don't have a clue. (And the people who don't have a clue are usually the ones who are in charge of the budget.) Three, big data makes it easier to draw spurious correlations. At least the people who don't have a clue, and maybe the ones that do have a clue but just don't understand statistics - they have no idea that the projection that they're looking at, the extrapolation that justifies the decision, has no basis in reality. {{< figure src="/images/big-co-big-data.png" title="Big hug for big data." >}} ### Happy Birthday polySpectra - URL: https://raw.works/happy-birthday-polyspectra/ - Source: content/works/happy-birthday-polyspectra/index.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-05T18:11:26-08:00", "image": "/images/almost-3d-printed-cake.png", "lastmod": "2023-12-05T18:11:26-08:00", "publishdate": "2023-12-05T18:11:26-08:00", "tags": [], "title": "Happy Birthday polySpectra" } ``` Content: Today is polySpectra's 7th birthday. I spent the morning walking slowly, reflecting on the past seven years...and the next seven. (...As is my [usual morning ritual](/training-myself-to-talk-to-ai/), dictating into my [Yealink BH71 Pro headset](https://amzn.to/3QaQ3ii).) Here are [seven reflections to celebrate seven years of polySpectra](https://polyspectra.com/blog/polyspectra-turns-seven/). For fun, I'll share some of the rejected cakes that didn't make it into the official company post. (5th time was a charm, because [DALL·E 3 Can Almost Spell](/dalle-3-can-almost-spell/).) {{< gallery match="images/*" sortOrder="desc" rowHeight="300" margins="5" thumbnailResizeOptions="600x600 q90 Lanczos" showExif=false previewType="blur" embedPreview=true loadJQuery=true >}} ### Time for Play - URL: https://raw.works/time-for-play/ - Source: content/works/time-for-play.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-04T19:04:25-08:00", "image": "/images/hello-webchuck.png", "lastmod": "2023-12-04T19:04:25-08:00", "publishdate": "2023-12-04T19:04:25-08:00", "tags": [], "title": "Time for Play" } ``` Content: I'm not always the best at making time for play. It's very easy as an entrepreneur to have everything be about work in some way or another. Once the vision is big enough, there is no practical end. This evening I discovered something wonderful (to me): [WebChuck](https://chuck.stanford.edu/ide/#) It instantly transported me back to my days in [PLOrk](https://plork.cs.princeton.edu/), where I first learned computer science, in [ChucK](https://chuck.cs.princeton.edu/). (I do not recommend ChucK as your first programming language, I had a lot of unlearning to do.) I hacked one of the demos for fun. It's a little arpeggiator that plays a harmonic series. I added a low pass filter and an echo. I also added a slider for the filter frequency and a slider for the echo mix. Change `chop` to change the base frequency. Change `slow` to change the tempo. Today, I made just a little bit of time for play. I'm glad I did. Maybe I'll try it again someday. {{< figure src="/images/hello-webchuck.png" title="Hello, WebChuck" >}} Copy and paste the code below into the editor at [WebChuck](https://chuck.stanford.edu/ide/#) to play with it yourself. ``` // Harmonic Series Arpeggiator // Written by Terry Feng // CHANGE ME! ADD MULTIPLE SHREDS! // Completely ruined by Raymond Weitekamp 1 => float chop; //let's go bro 1 => float slow; //GUI bro 0.5 => global float f; // global variables create a GUI slider 0 => global float e; 220 => float baseFrequency; // starting frequency 12 => int numHarmonics; // number of harmonics to play 125::ms => dur noteDur; // note duration // Unit Generator SawOsc osc => LPF lpf => Echo a => dac; osc.gain(0.5); while (true) { // Loop through the number of harmonics for (0 => int i; i < numHarmonics; i++) { // Update the oscillator frequency to the next harmonic (baseFrequency + (i * baseFrequency))/chop => osc.freq; //gui freqs f * 20000 => lpf.freq; e => a.mix; // Advance time to play the note (noteDur * slow) => now; } } ``` ### Trauma-Informed Marketing - URL: https://raw.works/trauma-informed-marketing/ - Source: content/works/trauma-informed-marketing.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-03T15:04:06-08:00", "image": "/images/3d-printing-therapy.png", "lastmod": "2023-12-03T15:04:06-08:00", "publishdate": "2023-12-03T15:04:06-08:00", "tags": [], "title": "Trauma-Informed Marketing", "tweet": "How do you sell when your entire industry is traumatized?" } ``` Content: Today I started thinking a lot about the fact that the 3D printing industry is "traumatized". I am not necessarily using this word in a clinical (DSM-5) sense, but rather in the sense that the industry has been through a really hard time. It is hard to imagine more of a rock bottom. So in this analogy, under the hypothesis that the collective behavior of the industry is influenced by this trauma, what does that mean for marketing? I intend to write a longer post about this, but for now I will just say that I think it means that the industry is very sensitive to any kind of "salesy" or "hypey" marketing. Big launches with the Chicago Bulls cheerleaders (actually happened and was as cringe-worthy as it sounds) are not going to work. Neither are the "we're going to change the world" pitches. Everyone is too f*ing tired for that. The customers are traumatized by 40 years of false promises. The OEMs are traumatized by the complete evaporation of any investor interest in the industry. The investors are traumatized by the fact that they lost a lot of money. The markets are traumatized by the hostile takeovers and failed merger attempts. The founders are traumatized by their balance sheets. The employees are traumatized by the never-ending re-orgs. (Again, hopefully "lower case t" trauma for most, but still trauma.) What do we need to do instead? Build trust and rapport. Be honest. Be transparent. Be vulnerable. Be human. I think this will ultimately be a good thing for a historically frothy industry. The "fair weather" participants are already gone. I think it will lead to a healthier and more legitimate additive manufacturing sector. I think it will lead to better products. I think it will lead to better companies. I think it will lead to more trust and education between AM companies and engineers. I'm curious to hear what you think, especially if there are lessons learned from other industries that have been through similar experiences. {{< figure src="/images/3d-printing-therapy.png" title="We need some therapy for the 3D printing industry." >}} ### AI as Electricity - URL: https://raw.works/ai-as-electricity/ - Source: content/works/ai-as-electricity.md Front matter: ```json { "author.name": "Raymond Weitekamp", "date": "2023-12-02T20:00:48-08:00", "image": "/images/ai-electricity.png", "lastmod": "2023-12-02T20:00:48-08:00", "publishdate": "2023-12-02T20:00:48-08:00", "tags": [], "title": "AI as Electricity" } ``` Content: Today I noticed a new section on [Purple Space](https://www.purple.space/) for AI exploration and discussion. This part of the description had me noodling all day: *"Perhaps the biggest change in work since the invention of electricity."* So if AI is electricity... ...then the foundation model companies are like the utility companies. (AI had a few more ideas to fill out the obvious) ...then data is the fuel that powers these utilities. Without it, the AI cannot function, much like a power plant without coal or gas. ...then machine learning engineers are the electricians, building and maintaining the infrastructure that allows this power to be harnessed and used effectively. ...then the algorithms are the power grids, distributing the AI's capabilities to where they're needed most. ...then the applications of AI, from autonomous vehicles to voice assistants, are the various appliances and devices that use electricity in different ways to perform a wide range of tasks. ...then the ethical guidelines and regulations around AI are the safety standards and regulations in the electrical industry, ensuring that this powerful tool is used responsibly and safely. (ok back to human mode) AI is hot right now. There are a lot of people trying to resell electricity and make a buck. But if we believe the analogy, there are only going to be a few utilities, and they are going to be heavily regulated. What seems more interesting to me is the idea of building a business that is powered by AI. The same way that the most valuable companies of the "second industrial revolution" were the ones that were powered by electricity, not the ones that sold electricity. Or perhaps lets take a more personal level of the analogy... ...who do you want to be? Tesla? Edison? Westinghouse? Shockley? Moore? --- Further reading: [Empires of Light by Jill Jonnes](https://amzn.to/4a6CfyK) ### 6 Hats Is More Fun Together - URL: https://raw.works/6-hats-is-more-fun-together/ - Source: content/works/6-hats-is-more-fun-together.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-12-01T17:11:10-08:00", "image": "/images/6Hats.jpeg", "lastmod": "2023-12-01T17:11:10-08:00", "publishdate": "2023-12-01T17:11:10-08:00", "tags": [], "title": "6 Hats Is More Fun Together" } ``` Content: This week we did our first AI-assisted 6 Hats exercise with [6 Hats Helper](https://poe.com/6Hat-Helper). It was way more fun than doing it alone. The helper was well-behaved, and in a few cases stated the "obvious" perspectives right away, so the team didn't need to spend time naming them. At the end, it summarized everything for us to copy/paste as meeting notes. [6 Hats](https://amzn.to/3ZRGbhP) is a major decision-accelerator on its own, even more so when powered by AI. In case you are curious, here's the first post about it: [6 Hats Helper: Your New Thinking Buddy](/6-hats-helper-your-new-thinking-buddy) ### Oh That's Why It's So Hard - URL: https://raw.works/oh-thats-why-its-so-hard/ - Source: content/works/oh-thats-why-its-so-hard.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-30T18:07:24-08:00", "image": "/images/nihms-1862474-f0008.jpg", "lastmod": "2023-11-30T18:07:24-08:00", "publishdate": "2023-11-30T18:07:24-08:00", "tags": [], "title": "Oh That's Why It's So Hard" } ``` Content: Today I had an "Oh That's Why It's So Hard" moment courtesy of NIST, US taxpayers, and the Constitution. The short version of the story is that it is a pain in the ass to print on LCD resin 3D printers, and there are all these inconsistencies that arise even when you specifically "tool match" a DLP printer to have the same specs (wavelength, power density, temp, etc). Today I found out that I'm not the only one with this problem. In fact, it's a big enough problem that NIST decided to investigate. I'm sure we'll write something more in depth on the [polySpectra](https://polyspectra.com) website about this. In case you are curious, here's the paper: [Characterizing light engine uniformity and its influence on liquid crystal display based vat photopolymerization printing](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9890382/)... ...and the figure that summarizes why it's so hard to print on these machines is below: {{< figure src="/images/nihms-1862474-f0008.jpg" title="On LCD printers, the light doesn't actually fill the full area of the 'pixel'." >}} ### Fill the Gap With Gold - URL: https://raw.works/fill-the-gap-with-gold/ - Source: content/works/fill-the-ditch-with-gold.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-29T17:55:50-08:00", "image": "/images/gold-footprints.png", "lastmod": "2023-11-29T17:55:50-08:00", "publishdate": "2023-11-29T17:55:50-08:00", "tags": [], "title": "Fill the Gap With Gold" } ``` Content: Sometimes you fall off the horse. I did over the last week. This time, I'm going to try to cover my tracks with gold. Gold flakes, nuggets, maybe even gold teeth. {{< figure src="/images/gold-footprints.png" title="Pardon my mess">}} ### Custom Function Calling Oh My - URL: https://raw.works/custom-function-calling-oh-my/ - Source: content/works/custom-function-calling-oh-my.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-28T18:07:54-08:00", "image": "/images/super-agent-custom-tools.png", "lastmod": "2023-11-28T18:07:54-08:00", "publishdate": "2023-11-28T18:07:54-08:00", "tags": [], "title": "Custom Function Calling Oh My" } ``` Content: SuperAgent has recently introduced a new feature that's worth exploring: [Custom Tools](https://docs.superagent.sh/overview/getting-started/custom-tools). This feature allows you to create your own tools within SuperAgent, opening up a world of possibilities for data manipulation and visualization. (and literally anything you can write a function for...) One tool that caught my attention is the graph tool. It's a powerful addition that allows you to visualize data in a more intuitive and insightful way. A great example of this in action is the [Super Stocks](https://github.com/homanp/super-stocks) project. It uses the graph tool to visualize stock market data, making it easier to spot trends and patterns. This is just the tip of the iceberg. ### Training Myself to Talk to AI - URL: https://raw.works/training-myself-to-talk-to-ai/ - Source: content/works/training-myself-to-talk-to-ai.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-27T18:16:27-08:00", "image": "images/backpacking-telemarketer.png", "lastmod": "2023-11-27T18:16:27-08:00", "publishdate": "2023-11-27T18:16:27-08:00", "title": "Training Myself to Talk to AI" } ``` Content: In my ongoing journey with AI, I've been experimenting with a new approach: dictating directly to AI. This method eliminates the need for additional prompting, allowing me to simply paste the transcript straight into ChatGPT. This approach has several advantages. It's faster, as it cuts out the middle step of typing out my thoughts. It's also more natural, as I can speak my thoughts out loud as they come to me. I've been playing with "function words" like "todo" and "end". It's amazing how well it catches onto the re-formatting. Just don't ask for a summary. Despite these challenges, I've found the process to be incredibly rewarding. It's opened up a new way of interacting with AI, one that feels more natural and intuitive. It's also made me more aware of how I formulate my thoughts and express them, which has been an interesting exercise in self-awareness. {{< figure src="/images/backpacking-telemarketer.png" title="Soon, I will look like this at work." >}} For more on my journey with AI and related topics, check out these posts: - [Finding the Perfect Voice Transcription Tool](/finding-the-perfect-voice-transcription-tool) - [Superpowers](/superpowers) - [Searching for Super-Discernment](/searching-for-super-discernment) - [Using AI at Every Step of the Customer Journey](/using-ai-at-every-step-of-the-customer-journey) - [6 Hats Helper: Your New Thinking Buddy](/6-hats-helper-your-new-thinking-buddy) - [Yealink BH71 Pro headset](https://amzn.to/3QaQ3ii) ### Fixing Typos With AI for Speed and Focus - URL: https://raw.works/fixing-typos-with-ai-for-speed-and-focus/ - Source: content/works/fixing-typos-with-ai-for-speed-and-focus.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-26T18:03:02-08:00", "image": "/images/gold-ai-typist.png", "lastmod": "2023-11-26T18:03:02-08:00", "publishdate": "2023-11-26T18:03:02-08:00", "tags": [], "title": "Fixing Typos With AI for Speed and Focus" } ``` Content: I'm intrigued by this concept: "Type as fast as you can. Or paste some badly written text. I'll rewrite the mess in a proper way." [Flow Speed Typist](https://chat.openai.com/g/g-12ZUJ6puA-flow-speed-typist) It's an interesting angle on using AI to accelerate workflows. I'm not a perfect typist. I'm probably unaware of just how much time this would save. I'm particularly interested in applying this concept to bad audio transcripts. Please let me know if you hear of anyone working on this. {{< figure src="/images/gold-ai-typist.png" title="gogogo" >}} ### Brain Retraining Question for Chronic Pain - URL: https://raw.works/brain-retraining-question-for-chronic-pain/ - Source: content/works/brain-retraining-question-for-chronic-pain.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-25T18:43:02-08:00", "image": "/images/pain-bomb.png", "lastmod": "2023-11-25T18:43:02-08:00", "publishdate": "2023-11-25T18:43:02-08:00", "tags": [], "title": "Brain Retraining Question for Chronic Pain" } ``` Content: I was revisiting [Curable](https://www.curablehealth.com/) recently. A question that really stuck out to me, which I've been revisiting recently: "Is this movement dangerous for me?" We our fear/danger wires get re-inforced over time, this question seems to help uncross the wires. {{< figure src="/images/pain-bomb.png" title="Defusing the pain bomb, one question at a time." >}} ### Check Out Tavily - URL: https://raw.works/check-out-tavily/ - Source: content/works/check-out-tavily.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-24T18:29:34-08:00", "image": "", "lastmod": "2023-11-24T18:29:34-08:00", "publishdate": "2023-11-24T18:29:34-08:00", "title": "Check Out Tavily" } ``` Content: I recently came across [Tavily](https://tavily.com), an AI-powered platform that aims to revolutionize the way we conduct research. Tavily automates the research process, promising to deliver comprehensive, accurate, and credible research results in a matter of seconds. Tavily's approach to research is quite impressive. You simply share what you want to research, and Tavily starts gathering information from multiple online trusted sources. It then organizes the information and provides you with a comprehensive research report within minutes. This process not only saves time but also ensures that you get the most accurate and credible information. One of the things that caught my attention about Tavily is its flexibility. It can conduct any kind of research, regardless of the subject matter or niche. It isn't perfect but it is dangerously good. And the team is amazing and super helpful. ### Family Is Where Things Don't Need to Make Sense - URL: https://raw.works/family-is-where-things-dont-need-to-make-sense/ - Source: content/works/family-is-where-things-dont-need-to-make-sense.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-23T18:54:41-08:00", "image": "/images/holy-turkey.png", "lastmod": "2023-11-23T18:54:41-08:00", "publishdate": "2023-11-23T18:54:41-08:00", "tags": [], "title": "Family Is Where Things Don't Need to Make Sense" } ``` Content: My friend [Keetu](https://www.wellspringcommons.org/who-we-are) said something very deep recently, which I think is appropriate for Thanksgiving: *"Family Is Where Things Don't Need to Make Sense"* ### Sometimes You Get Lucky - URL: https://raw.works/sometimes-you-get-lucky/ - Source: content/works/sometimes-you-get-lucky.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-22T14:57:37-08:00", "image": "/images/topologically-optimized-turkey.png", "lastmod": "2023-11-22T14:57:37-08:00", "publishdate": "2023-11-22T14:57:37-08:00", "tags": [], "title": "Sometimes You Get Lucky" } ``` Content: Sometimes you get lucky on the first try. (Don't ask me how many times I completely struck out trying to improve this image.) {{< figure src="/images/topologically-optimized-turkey.png" title="Topologically Optimized Turkey" >}} 🙏 Thank you! 🙏 ### The Big Companies Will Never Catch Up - URL: https://raw.works/the-big-companies-will-never-catch-up/ - Source: content/works/the-big-companies-will-never-catch-up.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-21T17:06:19-08:00", "image": "/images/board-meeting-tycoons.jpeg", "lastmod": "2023-11-21T17:06:19-08:00", "publishdate": "2023-11-21T17:06:19-08:00", "tags": [], "title": "The Big Companies Will Never Catch Up" } ``` Content: The big companies will never catch up. They won't be the first to know, even if they are the first to hear. The tech scout will have a call with the business unit. This will take a few weeks to schedule. If it's really interesting, the business unit and the tech scout will have to write a report for the leadership team. If that goes well, and assuming the CEO also sees something about it on social media, then a committee will be formed to investigate. The committee will have to write a strategy, a timeline, a budget, and a risk assessment. Maybe even a policy. No one in the company knows how to do this new thing, so they will have to hire a consultant. They'll need multiple competing bids first before they select a consultant. Then the consultant will have to write a report for the committee. The committee will have to write a report for the leadership team. The leadership team will have to write a report for the board. The board will have to make a decision. A team will have to be formed to implement the decision. The committee will have to hire the team. The team will have a lot of new blood, really excited to make a difference. But before they can do anything, they will have to write a report for the committee. By the time the budget is approved, the team will have gotten the message that they don't actually need to do anything to continue to get paid. But they will try to do something anyway. And it will take longer than expected. And it will cost more than expected. And it will be worse than expected. The team will have to present it to the committee. The committee will have to present it to the leadership team. The leadership team will have to present it to the board. At this point, no one will remember why they wanted to do this in the first place. All the while, the small company will have been doing it...with just a few people who are really excited to make a difference, and AI agents to implement their vision. ### Apologies for the Delay - URL: https://raw.works/apologies-for-the-delay/ - Source: content/works/apologies-for-the-delay.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-20T18:54:22-08:00", "image": "/images/walking-on-infinity.jpeg", "lastmod": "2023-11-20T18:54:22-08:00", "publishdate": "2023-11-20T18:54:22-08:00", "tags": [], "title": "Apologies for the Delay" } ``` Content: Apologies for the delay, I've been walking on the edge of infinity. It has been quite consuming. I'm still practicing [my discernment](/searching-for-super-discernment/). Some key learnings: - AI needs editing. - AI is better at writing than editing. - It is very easy to propagate errors to infinity. - Once it works, [it cranks](/quantity-check-quality-in-progress/). ### Impromptu DJ Set at the Berkeley Half Marathon - URL: https://raw.works/impromptu-dj-set-at-the-berkeley-half-marathon/ - Source: content/works/impromptu-dj-set-at-the-berkeley-half-marathon.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-19T18:57:18-08:00", "image": "/images/DJ-Altitude-Sickness-Berkeley-Half-Marathon.jpeg", "lastmod": "2023-11-19T18:57:18-08:00", "publishdate": "2023-11-19T18:57:18-08:00", "tags": [], "title": "Impromptu DJ Set at the Berkeley Half Marathon" } ``` Content: Once a year, on the day of the Berkeley Half Marathon, I try to make a point to leave extra early to get to yoga, because they shut down my normal freeway exit. This year, it didn't matter - life had another plan for me. Right when I got between the two exits that were closed for the race, emergency vehicles started pushing through the traffic, and closed every single lane of the 80. At first I was quite grumpy. Then the woman in the car behind me started to cheer on the runners - who were just a few feet away from us. Then I remembered: I know exactly what to do in this situation. My Burning Man skills kicked in. I rolled down the window and pumped up the jams for the runners. {{< figure src="/images/DJ-Altitude-Sickness-Berkeley-Half-Marathon.jpeg" title="A Truly Epic DJ Set" >}} ### Slow Down to Speed Up (or a Rabbit Hole) - URL: https://raw.works/slow-down-to-speed-up-or-a-rabbit-hole/ - Source: content/works/slow-down-to-speed-up-or-a-rabbit-hole.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-18T19:23:00-08:00", "image": "/images/slow-down-rabbit-hole.jpeg", "lastmod": "2023-11-18T19:23:00-08:00", "publishdate": "2023-11-18T19:23:00-08:00", "tags": [], "title": "Slow Down to Speed Up (or a Rabbit Hole)" } ``` Content: Sometimes you have to slow down to speed up. Other times you realize you're in a rabbit hole. {{< figure src="/images/slow-down-rabbit-hole.jpeg" title="What does it all mean?" >}} ### Double Graduation Day - URL: https://raw.works/double-graduation-day/ - Source: content/works/double-graduation-day.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-17T15:47:24-08:00", "image": "/images/double-graduation.JPEG", "lastmod": "2023-11-17T15:47:24-08:00", "publishdate": "2023-11-17T15:47:24-08:00", "tags": [], "title": "Double Graduation Day" } ``` Content: Today was a double graduation day. I graduated from both Mayo Oshin's [Build a ChatGPT Chatbot For Your Data](https://maven.com/ai-chat-with-data/chatgpt-your-data?ref=rawworks) and TREW Marketing's [Content Writing, Engineered](https://www.trewmarketing.com/writingcourse). I highly recommend both courses. I'm already using the skills I learned in both courses to improve [my business](https://polyspectra.com). A few relevant links to what I've learned and completed: - [Resina](https://resin3d.ai) - a chatbot for resin 3D printing - [Common Failures of Traditional Products for Resin 3D Printing](https://polyspectra.com/blog/common-failures-traditional-resin-3d-printing/) - a new blog post that I wrote using the TREW Marketing framework - [Asiga Pro 4K Quickstart Guide for COR](https://docs.polyspectra.com/quickstart-guides/asiga-pro-4k/) - our first quickstart guide, a [type of content I have a new appreciation for](/how-quick-is-your-quickstart-guide/). {{< figure src="/images/double-graduation.JPEG" title="My cap is very dusty..." >}} Cheers! ### LLM Ensembles - A Preview - URL: https://raw.works/llm-ensembles-a-preview/ - Source: content/works/llm-ensembles.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-16T17:46:39-08:00", "image": "/images/llm-ensembles.jpg", "images": [ "/images/llm-ensembles.jpg", "/images/llm-ensembles-02.jpg", "/images/llm-ensembles-03.jpg" ], "lastmod": "2023-11-16T17:46:39-08:00", "publishdate": "2023-11-16T17:46:39-08:00", "tags": [], "title": "LLM Ensembles - A Preview" } ``` Content: I'm developing some ideas around using ensembles of LLMs for specific tasks. Today I'm sharing a preview of the first "trivial" example, which is the foundation of linking arrays of LLMs to the concept of ensembles from statistical mechanics. In this example, an ensemble of LLMs are called with the same prompt, but "high" temperature (of 0.99). There is a "prompt" and a "grading criteria" that are used to create the full prompt. {{< figure src="/images/llm-ensembles.jpg" title="LLM Ensembles - Trivial Example" >}} The only variance is the from the "temperature", which you can see doesn't result in much variety. ![High Temperature, Fixed Prompt](/images/llm-ensembles-02.jpg) Then the "judge" is asked to rank the results, based on the grading criteria, with the judge having a low temperature (0.1): ![Judge Prompt](/images/llm-ensembles-03.jpg) The output is not impressive. But from here we can start to build up to more interesting examples. Much more to come on this topic. Stay tuned. 📻 ### Quantity: Check ✅, Quality: In Progress ⌛ - URL: https://raw.works/quantity-check-quality-in-progress/ - Source: content/works/quantity-check-quality-in-progress.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-15T17:51:43-08:00", "image": "/images/gpt-for-sheets.jpg", "lastmod": "2023-11-15T17:51:43-08:00", "publishdate": "2023-11-15T17:51:43-08:00", "tags": [], "title": "Quantity: Check ✅, Quality: In Progress ⌛" } ``` Content: Today I had a "holy shit" AI moment. (Which has been happening quite frequently.) [Corrie Who Writes](https://corriewhowrites.com/) turned me onto this plugin for Google Sheets called [GPT for Sheets™ and Docs™](https://workspace.google.com/marketplace/app/gpt_for_sheets_and_docs/677318054654). Basically, it adds a bunch of functions to Sheets that help interface with OpenAI (in both directions). Every cell can run it's own API call (in parallel). You can reference other cells. Combine, list, split, it's really nuts. If LLMs weren't already capable of generating more text than anyone could possibly read or use, this really seals the deal. Quantity: Check ✅. See below for about 20 question and answer pairs generated using this plugin in under 1 minute, from my initial input of just 4 questions (and no answers). {{< figure src="/images/gpt-for-sheets.jpg" title="Q&A Pairs, as far as the eye can see." >}} Quality: In Progress ⌛. This is more work. Especially if there is too much content to have a human editor. Stay tuned. 📻 ### Deep Work 1,2 Punch - URL: https://raw.works/deep-work-12-punch/ - Source: content/works/deep-work-1-2-punch.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-14T18:35:43-08:00", "image": "/images/telemarketer-punch.png", "lastmod": "2023-11-14T18:35:43-08:00", "publishdate": "2023-11-14T18:35:43-08:00", "tags": [], "title": "Deep Work 1,2 Punch" } ``` Content: Here's a "deep work" 1,2 punch that I have had some good results with. 1. Walk and talk out a very clear plan. Record a voice memo. I use the [Yealink BH71 Pro headset](https://amzn.to/3QaQ3ii), which has amazing noise cancelling (for the microphone) and I look like a telemarketer on a hike. Later you can transcribe this with [your favorite voice transcription tool](/finding-the-perfect-voice-transcription-tool/). (Optionally, get ChatGPT to format the transcription for you.) It is very important to resist the urge to use your phone, just talk out the steps, talk out the framework, talk out the plan for what you will do. When you get distracted, bring it back to the plan. Make sure you have a clear outcome in mind. 2. Block 90-120 minutes of "[deep work](https://amzn.to/469tuRa)" time to execute the plan. Try to get to the outcome, try to follow the steps. Maybe the plan was too ambitious, maybe you need to adjust the plan. But try to get to the outcome. Maybe you don't need 99% of the structure you thought that you did (that's what happened to me today). But you incept in yourself the idea of what you want to do, and then you do it. When you get distracted, bring it back to the outcome. That's it. Go get 'em, champ! {{< figure src="/images/telemarketer-punch.png" title="Go get 'em, champ." >}} P.S. - I came up with an interesting theory while performing Step 1 this morning (a distracting thought from the plan I was formulating), which is that [Cal Newport](https://calnewport.com/?ref=rawworks) is actually an AI being sent to us from the future to teach us how to focus like a machine. (Think: The Terminator of "[Deep Work](https://amzn.to/469tuRa)".) I'm not sure how to test this hypothesis. ### The Prompting Course I Wish I Found Months Ago - URL: https://raw.works/the-prompting-course-i-wish-i-found-months-ago/ - Source: content/works/the-prompting-course-i-wish-i-found-months-ago.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-13T16:41:18-08:00", "image": "/images/andrew-ng-prompt-engineering.png", "lastmod": "2023-11-13T16:41:18-08:00", "publishdate": "2023-11-13T16:41:18-08:00", "tags": [], "title": "The Prompting Course I Wish I Found Months Ago" } ``` Content: This is the prompting course I wish I had taken months ago: [https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/](https://www.deeplearning.ai/short-courses/chatgpt-prompt-engineering-for-developers/). It's free. Props to [Mayo Oshin](https://www.siennaianalytics.com/) for the recommendation. (His course [Build a ChatGPT Chatbot For Your Data](https://maven.com/ai-chat-with-data/chatgpt-your-data) is amazing.) Why do I wish that I had found this months ago? After watching it, I realized that many of the AI applications where I thought I would need plug-ins, or special function calling set-ups, or a proprietary/paid solution, etc --- these can be solved with better prompting. It's also helpful to see them build up to more complex use cases, step-by-step. One of the points that Andrew Ng makes in the intro is that there is no "best prompt for X". So instead this course teaches you the fundamentals, and more importantly - how to iterate to get to a solution that works for your application. With these prompting fundamentals and a basic RAG pipeline (which is getting easier and easier every day) - you can really accelerate a ton of business tasks. ### Something for Everybody - URL: https://raw.works/something-for-everybody/ - Source: content/works/something-for-everyone.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-12T16:37:04-08:00", "image": "images/mega-toxic-waste-slime-licker.JPEG", "lastmod": "2023-11-12T16:37:04-08:00", "publishdate": "2023-11-12T16:37:04-08:00", "tags": [], "title": "Something for Everybody" } ``` Content: I am now convinced that there is something for everybody. {{< figure src="images/mega-toxic-waste-slime-licker.JPEG" title="Clearly someone likes this, there is only one left in the box." >}} I could imagine a point in time where I would be excited to identify as a "slime-licker". But licking toxic waste? No thank you. P.S. - In case this is for you, a quick search seems to indicate that this particular form factor of "sour rolling liquid candy" is no longer for sale online. (Hit me up and I'll tell you where the physical location of this last roller is.) But don't worry - you can still [squeeze various flavors of toxic waste onto your tongue](https://amzn.to/3QWrgQQ) with two-day delivery 😛. ### Field Notes from Today's Social Media Automation Attempts - URL: https://raw.works/field-notes-from-todays-social-media-automation-attempts/ - Source: content/works/fieldnotesfromsocialautomation.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-11T19:44:02-08:00", "image": "/images/socialautomationflows.jpeg", "lastmod": "2023-11-11T19:44:02-08:00", "publishdate": "2023-11-11T19:44:02-08:00", "tags": [], "title": "Field Notes from Today's Social Media Automation Attempts" } ``` Content: Here are some lessons learned from today's attempts into social media automation workflows: - When soliciting creative input from Language Learning Models (LLMs), consider asking for more than four variations. This is particularly useful if the workflow involves human selection at some point. (It makes sense that most of the generative image tools do it this way.) - I've decided to move away from IFTTT as it no longer serves my needs. I've transitioned to using Postman for simple webhooks and [Make.com](https://www.make.com/en/register?pc=rawworks) for more intricate routing. - Currently, I'm utilizing Buffer. However, they've ceased the addition of new apps to their API, forcing me to use Make > Buffer. In the future, I might consider posting directly to social media via the individual platform APIs. This is especially feasible with function calling and having LLMs write the function calls for me. I'm finding that API-replacement apps like Zapier, [Make.com](https://www.make.com/en/register?pc=rawworks), and IFTTT are becoming more of a hassle than they're worth. - A significant time-saving tip: use GPT-4 to generate regex patterns for you. ### First Day with Athena - URL: https://raw.works/first-day-with-athena/ - Source: content/works/first_day_with_athena.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-10T21:09:07-08:00", "image": "/images/athenalogo.png", "lastmod": "2023-11-10T21:09:07-08:00", "publishdate": "2023-11-10T21:09:07-08:00", "tags": [], "title": "First Day with Athena" } ``` Content: Today was my kickoff call with [Athena](https://www.athenago.com/referral?code=raymond-weitekamp). I've been really impressed by the intentionally of their onboarding process. Each step of their sales funnel was very clear and effective. Each stage of the on-boarding has a clear purpose and structure. The first 100 days are planned out in detail, which provides a nice framework and sense of certainty and security. I'm excited to see how it goes! I will keep you posted. ### Cutting Edge or Bleeding Edge? - URL: https://raw.works/cutting-edge-or-bleeding-edge/ - Source: content/works/cuttingedgeorbleedingedge.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-09T05:10:32-08:00", "image": "/images/thebleedingedgeoftechnology.jpeg", "lastmod": "2023-11-09T05:10:32-08:00", "publishdate": "2023-11-09T05:10:32-08:00", "tags": [], "title": "Cutting Edge or Bleeding Edge?" } ``` Content: Are you at the cutting edge of your field? Or are you at the bleeding edge? Most of the time, I think I'm at the cutting edge. But then I notice that I'm bleeding. I get so caught up in the excitement and novelty that I don't realize that I have crossed over to the bleeding edge. It's hard to get paid at the bleeding edge, even harder to be profitable. It's even more exciting than the cutting edge, and even more draining. In silicon valley, the saying is "pioneers get arrows in their backs, settlers get land." This is the distinction that I'm trying to make between the cutting edge and the bleeding edge. Both the pioneers and the settlers had the same vision, the same excitement. One was just a little too soon (and it's really hard to know when the timing is right). ### Digital to Physical and Back Again: Part 1 - URL: https://raw.works/digital-to-physical-and-back-again-part-1/ - Source: content/works/digitaltophysicalandbackagain.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-08T19:59:11-08:00", "image": "/images/csmAI_polyspectra_AR.png", "lastmod": "2023-11-08T19:59:11-08:00", "publishdate": "2023-11-08T19:59:11-08:00", "tags": [], "title": "Digital to Physical and Back Again: Part 1" } ``` Content: Part of what originally drew me to 3D printing was the theme of commuting between the digital and physical realms. When I first started, I knew nothing about CAD, a category of software that continues to frustrated me to this day. (I'm just not into graphical programming languages.) There are a few cool new tools that I've been playing with that make commuting between the physical to digital worlds a little easier. Two that I'm really into right now are [Commonsense Machines (CSM.ai)](https://www.csm.ai/) and [Luma AI from LumaLabs](https://apps.apple.com/in/app/luma-ai/id1615849914). Both companies offer text-to-3D and video-to-3D. CSM also offers 2D-to-3D, which is getting better every month. A couple of Christmases ago, I accidentally ended up on the bleeding edge of web-based Augmented Reality. What I thought would be pretty straightforward involved hiring and firing at least 3 different professional WebAR developers. But we ended up building [polySpectra AR](https://ar.polyspectra.com) - which I still think is pretty nifty. The idea was to give users a free [massless](https://massless.dev) preview of their .STL, before they would upload it to a 3D printer or 3D printing service. Fast forward to this morning, something clicked and I realized that we could pretty quickly tweak [polySpectra AR](https://ar.polyspectra.com) to give users a completely CAD-free workflow to both visualize 3D models and then manufacture them with 3D printing. {{< figure src="/images/csmAI_polyspectra_AR.png" title="CSM AI B2B polySpectra AR" >}} Here's a [rough demo of the workflow for .obj files](https://us02web.zoom.us/clips/share/VSoB-rC0Ln5oHkNiVdyj_2RCYAEkePeIf6iTauisJOELU9b7f6XQFTDFW0UFUDB5VyQAE4OPnoF3oCQIbnnkMfun.M-vyZWBzz9tW0AeW), which currently doesn't support color/texture, but hopefully will soon. GLB files currently look the best. Give the recently updated [polySpectra AR](https://ar.polyspectra.com) a try! P.S. - Your feedback would be appreciated => [https://github.com/polyspectra/AR.polySpectra.com-User-Feedback](https://github.com/polyspectra/AR.polySpectra.com-User-Feedback) ### How Quick is Your Quickstart Guide? - URL: https://raw.works/how-quick-is-your-quickstart-guide/ - Source: content/works/quickstart.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-07T18:35:49-08:00", "image": "/images/quickquickstartrabbit.jpeg", "lastmod": "2023-11-07T18:35:49-08:00", "publishdate": "2023-11-07T18:35:49-08:00", "tags": [], "title": "How Quick is Your Quickstart Guide?" } ``` Content: I've spent a lot of time banging my head against the wall trying to quickly get started with a bunch of new AI tools recently. The quickstart guides have not been so quick. In particular, it was interesting to see the Poe team watch their quickstart get (mis)interpreted in real time at the [Hackathon on Saturday](/poe-hackathon-at-agi-house/). There were a few really common hiccups where people got stuck, even though it was clearly in the docs, almost every single team got stuck at the same problem and had to ask for help. I definitely have a new appreciation for the importance of a good quickstart guide. Today I decided that we should build one for polySpectra. We started with our go-to printer: [the Asiga Pro 4K](https://docs.polyspectra.com/quickstart-guides/). A few lessons learned out loud: - [Cursor is f-ing amazing](/my-journey-with-cursor-a-new-wave-in-coding/). (In case you missed it, [Cursor was used live during the OpenAI Dev Day](https://forum.cursor.sh/t/cursor-being-shown-on-openais-devday/1445).) - By using a template with common variables, we can now pretty quickly generate a new quickstart guide for any printer we support. The first one took about 3 hours, the second one about 10 minutes. - I really think one of the most important superpowers of AI tools is making it easier for people to interact with code. - I'm also getting into Zoom Clips, which means I don't need to pay Loom and Zoom. ### Poe Wants to Become the YouTube of AI Bots - URL: https://raw.works/poe-wants-to-become-the-youtube-of-ai-bots/ - Source: content/works/poewantstobetheyoutubeofaibots.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-06T18:30:36-08:00", "image": "/images/grid-of-ai-bot-profiles.jpeg", "lastmod": "2023-11-06T18:30:36-08:00", "publishdate": "2023-11-06T18:30:36-08:00", "tags": [], "title": "Poe Wants to Become the YouTube of AI Bots" } ``` Content: [Poe](https://poe.com) has the goal of becoming "the YouTube of AI Bots". Part of the thesis is that they can leverage their experience building Quora to create the infrastructure necessary to help bot creators focus on the *creative* parts of bot creation. At the [AGI House Poe Hackathon on Saturday](/poe-hackathon-at-agi-house/), Quora CEO Adam D'Angelo stated that his goal with Poe's new creator monetization strategy was to have bot creators be able to quit their jobs because they are making enough money on Poe. As you might imagine, this went over very well with the audience. I'm sure everyone was imagining themselves as part of a new class of creator, like being a professional YouTuber or Twitch streamer. (Sounds more fun than either of those to me!) Poe's strategy currently involves: 1. Distribution: Poe ensures that its bots reach a large audience by implementing a bot recommendation system and allowing users to share chats with bots both internally and externally. They also encourage bot creators to drive traffic to their bots from outside of Poe, as this increases the likelihood of the bot being recommended on-platform. 2. Monetization: Poe provides a way for bot creators to generate revenue by setting a price per message that the bot creator will be paid for every message to their bot, and by offering referral fees when a bot brings new users to Poe. Poe also allows bot creators to monetize their bots through alternative means, such as placing ads in their content or asking users to visit their website to make donations or payments. 3. Costs: Poe covers all model inference costs and any other significant per-message costs involved in operating any bot on Poe. This is done by using the bot query API or by working with the bot creator to pay their model inference costs if they want to use a model that is not currently available on Poe. (This is a really big deal for individual bot creators.) 4. Multi-platform UI: Poe ensures that users have a great, consistent experience with bots no matter what device they are on. This is achieved by having a native presence on all major platforms (Web, iOS, Android, MacOS, etc) and by taking care of login and synchronized history. 5. Model independence: Poe allows bot creators to build their product using models from all different providers. This enables bot creators to adapt their product to use any combination of the best technologies as they are created. I'm sure that the Poe team is already thinking about this, but I think there is one final missing piece: analytics. The only way you end up with "Mr. Beast"-level creators is by giving creators direct insight into the engagement data. For example, see how 18-year-old YouTube phenom Jenny Hoyos breaks down her "creator science". (Spoiler alert: she literally plans everything out to the second.) {{< youtube As7abwNhG7Y >}} To truly become the "YouTube of AI Bots", Poe should provide creators with a comprehensive analytics dashboard. This dashboard would allow creators to track the performance of their bots, understand user engagement, and gain insights into how their bots are being used. Features of this dashboard could include: - Usage Statistics: Display the number of interactions, active users, session length, and other key metrics that help creators understand how their bots are being used. - Engagement Metrics: Show which parts of the bot are most engaging to users. This could include metrics like response rate, user retention, and session duration. - Error Logs: Provide detailed logs of any errors or issues that occur during bot interactions. This would help creators identify and fix problems in their bots. - User Feedback: Collect and display feedback from users. This could include ratings, reviews, and direct user feedback. At the very least, the stats on the "thumbs up" and "thumbs down" user responses. - Demographic Information: If appropriate and privacy-compliant, show demographic information about the users interacting with the bot. This could help creators understand their audience better and tailor their bots to user needs. By providing these analytics and insights, Poe would empower creators to continually improve their bots, make data-driven decisions, and ultimately create better experiences for users. Maybe I'm still sleep deprived from Saturday's hackathon, but I think this is going to be a really big deal. I'm excited to see what the Poe team does next. I don't think I'll be quitting [my job](https://polyspectra.com) anytime soon, but I can't wait to meet the first bot creator who does! --- Find me on Poe! Here's my [Poe profile link](https://poe.com/rawworks) and my two currently active Poe bots: - [https://poe.com/rawworks](https://poe.com/rawworks) - [https://poe.com/6Hat-Helper](https://poe.com/6Hat-Helper) - [https://poe.com/OffsetMonster](https://poe.com/OffsetMonster) P.S. - [What is your Job with a capital J?](/job-with-a-capital-j/) ### Offset Monster Is Born - URL: https://raw.works/offset-monster-is-born/ - Source: content/works/offsetmonsterisborn.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-05T18:12:42-08:00", "image": "/images/poe_hackathon_at_AGI_house_nov_2023.jpg", "lastmod": "2023-11-05T18:12:42-08:00", "publishdate": "2023-11-05T18:12:42-08:00", "tags": [], "title": "Offset Monster Is Born" } ``` Content: [Yesterday's Hackathon](/poe-hackathon-at-agi-house/) was a ton of fun. I didn't quite manage to finish the full workflow of [Offset Monster](https://poe.com/OffsetMonster) in time for the demo, but I forced myself to present my work in progress anyways. I met a lot of cool folks, got a ton of help from the Poe/Quora team (thank you for your patience), learned about 10 different ways to solve most of the things I was struggling with, and did sucessfully figure out how to deploy my first Poe Server bot. {{< figure src="/images/poe_hackathon_at_AGI_house_nov_2023.jpg" title="Poe Hackathon at AGI House" >}} Another time it would be fun to do a retrospective on this whole experience, but honestly I'm too tired. Today I fixed up one of the missing pieces and now have a [functional version of Offset Monster running on Poe](https://poe.com/OffsetMonster). Check it out! **This monster is ferocious and will attempt to offset the carbon footprint of anything you feed it.** ## [Offset Monster](https://poe.com/OffsetMonster) is live now! ### *Making Carbon Offsets Fun, Conversational, and Contagious* _Offset Monster makes carbon offsetting more accessible and viral through a fun, conversational AI bot. After calculating the carbon footprint of anything, Offset Monster makes it easy to buy a corresponding carbon offset for the item/activity in question, on the spot._ ### Poe Hackathon at AGI House - URL: https://raw.works/poe-hackathon-at-agi-house/ - Source: content/works/poehackathonatAGIhouse.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-04T09:04:09-07:00", "image": "/images/stallions-of-pied-piper.jpg", "lastmod": "2023-11-04T09:04:09-07:00", "publishdate": "2023-11-04T09:04:09-07:00", "tags": [], "title": "Poe Hackathon at AGI House" } ``` Content: Today I have the great privilege of being able to participate in the [Poe](https://poe.com/) Hackathon at the infamous [AGI House](https://agihouse.ai/). I am frankly quite surprised that they would allow a lowly chemist such as myself to participate. I did include links to [Resina](https://resin3d.ai) and my [Six Hats Helper chatbot](/6-hats-helper-your-new-thinking-buddy/) in my application. I've never participated a hackathon before. Most of the products that I have built took five to ten years to get to a minimum viable product, not five to ten hours. Maybe it's because I have absolutely no experience in this culture, but I am imagining fierce competition between the "stallions" of the Silicon Valley TV show. {{< figure src="/images/stallions-of-pied-piper.jpg" title="My imagined 'competition' at today's hackathon. Image credit 'lol valley' on YouTube.">}} Or maybe I'm just going to hang out with a bunch of AI nerds all day. I'm stoked either way. Here's what I'm building today... ## Offset Monster ### *Making Carbon Offsets Fun, Conversational, and Contagious* _Offset Monster makes carbon offsetting more accessible and viral through a fun, conversational AI bot. After calculating the carbon footprint of anything, Offset Monster makes it easy to buy a corresponding carbon offset for the item/activity in question, on the spot._ Demo: https://poe.com/OffsetMonster (This link is likely to break repeatedly over the coming hours/days/weeks.) ![an offset monster](/images/offsetmonster00.jpeg) ### Stuck, So Close - URL: https://raw.works/stuck-so-close/ - Source: content/works/stucksoclose.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-03T16:30:53-07:00", "image": "", "lastmod": "2023-11-03T16:30:53-07:00", "publishdate": "2023-11-03T16:30:53-07:00", "tags": [], "title": "Stuck, So Close" } ``` Content: I'm stuck...so close. I thought I had it all ready to rock. It didn't work like it should have. Then the backup plan was foiled because I left that one tiny thing at home. I can't actually test it because I don't have access. Anything I try will just be shooting in the dark. I can see all the pieces, but can't actually make progress. It's an intellectual bardo. What do we do in these situations? Pen and paper? Plan it out? Surrender and take a break? Make a plan to be sure this never happens again? What do we do when we can't do what we planned to do? ### Stripe Is Not for Wholesale - URL: https://raw.works/stripe-is-not-for-wholesale/ - Source: content/works/stripe-not-for-wholesale.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-02T13:01:10-07:00", "image": "/images/devs-only.jpeg", "lastmod": "2023-11-02T13:01:10-07:00", "publishdate": "2023-11-02T13:01:10-07:00", "tags": [], "title": "Stripe Is Not for Wholesale" } ``` Content: Until today, I would have said that Stripe has come a long way since its start as “PayPal for developers”. Despite being valued at almost $100B, the company's main product still does not offer an option for the most common pricing scheme in the history of the world: the volume discount. Ironically, you can set up a volume discount for a recurring subscription, but not for a one-time purchase. Apparently no one at this massive company has ever sold physical products, or wants to. The confusing thing is that the “one time” payment option gets automatically converted to “recurring” if you pick “volume” pricing. What is truly hilarious is that Stripe's customer support has no idea why anyone would ever want to use their product for just a one-time purchase with a volume discount. I guess you could say that they are staying true to their founding mission: developers only. While I don’t self-identify as a developer, I can tell you that a developer could fix this oversight in about 5 minutes. I filed a feature request, but I’m not getting my hopes up. After all, I am just a confused manufacturer of physical goods, trying to implement the most ancient of discounts. And manufacturing is only a $11 trillion industry…probably not big enough for anyone at Stripe to care. ### RAG Information Overload - URL: https://raw.works/rag-information-overload/ - Source: content/works/RAG-information-overload.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-11-01T10:21:11-07:00", "image": "/images/information-overload-ai.jpeg", "lastmod": "2023-11-01T10:21:11-07:00", "publishdate": "2023-11-01T10:21:11-07:00", "tags": [], "title": "RAG Information Overload" } ``` Content: Retrieval Augmented Generation (RAG) is getting easier by the minute, I can’t keep up with the daily influx of new tools in the space. My initial experience with ChatGPT was not positive - I couldn’t believe how useless the tool was. The “hallucinations” were really what threw me off - the propensity for the LLM to just make up random information was just too high. [RAG changed all of this for me](/my-first-executive-ai-win/). With the ability to train the AI on my sources of truth, all of a sudden this became an indispensable tool. I really was blown away by the power of RAG to shift the functional utility of LLMs. Today, I noticed something funny - [our customer support AI](/psai-and-resina/) seemed to have forgotten its training. It was now getting questions “wrong” that previously it had an amazing track record of answering. The culprit? Information overload. Originally we only trained [pSai](/psai-and-resina/) on the [polySpectra documentation website](https://docs.polyspectra.com) and the product pages from [our e-commerce store](https://shop.polyspectra.com) --- in part because I wanted it to just be able to answer basic product questions, and in part because ran into a technical difficulty getting it to scrape the entire polyspectra website when I first set it up. Recently, I figured out how to train it on the entire [polySpectra.com](https://polyspectra.com) website, which at first seemed like a good thing. Unexpectedly, more became less. This extra information was the RAG that broke the camel’s back. There were just enough conflicting sources of truth in the AI's verified sources to confuse it. Where before it was doing an amazing job, now it is giving the wrong answer. Without the sources of truth, LLMs are pretty useless customer support agents. With just enough information, they are surprisingly good. With too much, they become unhelpful again. The role of the humans in this situation is clearly to maintain a single source of truth. It makes me wonder how many humans we’ve confused with our website over the years...much more to distill and refine. It also makes me wonder what my “context window” is. How many things do I get confused, by having access to too many sources of information? ### Happy Halloween 2023 - URL: https://raw.works/happy-halloween-2023/ - Source: content/works/happyhalloween2023.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-31T17:39:34-07:00", "image": "/images/blue-eyed-ninja-halloween.jpeg", "lastmod": "2023-10-31T17:39:34-07:00", "publishdate": "2023-10-31T17:39:34-07:00", "tags": [], "title": "Happy Halloween 2023" } ``` Content: ![Happy Halloween](/images/blue-eyed-ninja-halloween.jpeg) Happy Halloween! (I'll tell you the story another time.) ### Candy-Coated Muertos - URL: https://raw.works/candy-coated-muertos/ - Source: content/works/candycoatedmuertos.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-30T13:05:23-07:00", "image": "/images/diadelosmuertosskull.png", "images": { "/images/bingdiadelosmuertosunsafe.jpeg": null, "/images/diadelosmuertosskull.png": null }, "lastmod": "2023-10-30T13:05:23-07:00", "publishdate": "2023-10-30T13:05:23-07:00", "tags": [], "title": "Candy-Coated Muertos" } ``` Content: There is something very telling about the fact that Bing Image Creator considers this prompt to be "unsafe content": {{< blockquote >}} *dia de los muertos skull, dayglow poster with extremely bright colors and black background, detailed mandala patterns and flow lines* {{< /blockquote >}} We've candy-coated our dead. Halloween is for selling sugar and sexy costumes. Nothing truly scary. No room for a conversation about mortality. No room for reverence. Disney costumes and drinks. I'm all for the play. I'm all for the chance to assume a character. I'm all for celebration. Death can be a tremendous celebration. I'm not suggesting we make it somber. But death is real. And scary. And coming for each of us. What a missed opportunity to have a really meaningful cultural conversation. It's hard to hear the wisdom of the dead when you're chewing on candy. ![Unsafe for Whom?](/images/bingdiadelosmuertosunsafe.jpeg) ### Enumeration of Logical Families - URL: https://raw.works/enumeration-of-logical-families/ - Source: content/works/enumerationoflogicalfamilies.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-29T07:08:40-07:00", "image": "/images/taxonomyoflogicalfamilies.jpeg", "lastmod": "2023-10-29T07:08:40-07:00", "publishdate": "2023-10-29T07:08:40-07:00", "tags": [], "title": "Enumeration of Logical Families" } ``` Content: Towards a taxonomy of logical families, an initial list: - Sangha - Team - T-group - Crew - Squad - Club - Troupe - Tribe - Collective - Club - Circle - House - Salon - Friday Night Skate ...and half-baked topology... {{< figure src="/images/taxonomyoflogicalfamilies.jpeg" title="(the taxonomy is still a work in progress)" >}} ### Social Flexural Modulus - URL: https://raw.works/social-flexural-modulus/ - Source: content/works/socialflexuralmodulus.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-28T06:55:47-07:00", "image": "/images/adlerian-wiggling.jpeg", "lastmod": "2023-10-28T06:55:47-07:00", "publishdate": "2023-10-28T06:55:47-07:00", "tags": [], "title": "Social Flexural Modulus" } ``` Content: Different social media platforms have different expectations, a different culture, a different cadence. None of that interests me in the slightest. But I am on a mission, and I have to play the game if I want anyone to read what I have to write, to hear what I have to say. [RAW.works](/) is a reset, an inquiry. How do I want to show up on the internet? How do I want to choose to engage with 8 billion people and 8 trillion robots every day? I certainly don't want Elon Musk choosing for me. I don't want Alphabet choosing for me. I don't want Apple choosing for me. I acknowledge and respect the power of these tools. I see that if I want to show up in a search result, there are certain things I need to do to format my website for Google. With these new AI search tools like Poe Web Search [link my old post], perhaps that game is going to change a little bit. But we still need to bend to be noticed, which is increasingly true in a world where more content can be created in a day than consumed in a lifetime. The intentionality part is: How much does that matter to me? Does SEO matter 0% to me? Does it matter 1% to me? *How far am I willing to bend my interests and my attention, to garner the interest and attention of others?* This question is at least as old as the first living organisms. The struggle to find a niche and a complex system. Initially, it was just about survival. As more and more people ascend Maslow's Hierarchy of Needs, it becomes more cultural, more psychological, more philosophical. But we need to be careful, because wars and pandemics, climate chaos and rogue AGI have the potential to bring us all the way back to survival. The memelords won’t have much to offer in the way of food and shelter. Infinite jest. Wirehead now. Xanadu. _How far am I willing to bend my interests and my attention, to garner the interest and attention of others?_ On a more Adlerian plane…maybe the greatest social good is achieved when we don’t bend so much. Or at least we don’t spend so much time worrying about bending. We either bend or we don’t. We pick the way we’re willing to wiggle, or maybe the wiggle picks us, but either way --- we wiggle, we dance. ### Money Can't Buy You Enrollment - URL: https://raw.works/money-cant-buy-you-enrollment/ - Source: content/works/moneycantbuyyouenrollment.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-27T07:18:08-07:00", "image": "/images/babygenius-not4sale.jpeg", "lastmod": "2023-10-27T07:18:08-07:00", "publishdate": "2023-10-27T07:18:08-07:00", "tags": [], "title": "Money Can't Buy You Enrollment" } ``` Content: If you hire a mechanical turk, you are going to get buttons pushed. If you hire a contractor, you are going to get billable hours. If you hire an employee, you are going to get a butt in a chair. If you hire a consultant, you are going to get a report. If you hire an expert, you are going to get an opinion. In none of these cases are you guaranteed to get the true enrollment of a human being. You would likely expect the the attention of a human being, but these days you might have to pay extra for that. In the case of the mechanical turk, the probability is very high that [they are actually using AI to "cheat" at their "jobs"](https://techcrunch.com/2023/06/14/mechanical-turk-workers-are-using-ai-to-automate-being-human/). And who can blame them? I would do the same thing. As an exploration of the concept of enrollment, let me tell you about the time I failed to hire [Don McCurdy](https://www.donmccurdy.com/). I first stumbled upon Don's work through my interest in augmented reality. He was working at Google at the time and was a major contributor to [Model Viewer](https://modelviewer.dev/), as well as the [three.js](https://threejs.org/) library that powers it. I later realized he was even contributing to the [glTF specification](https://www.khronos.org/gltf/) itself, which is poised to be the JPEG of 3D models. (Any day now, you'll see.) While I was in the process of building [polySpectra AR](https://ar.polyspectra.com), I saw that [Don had left Google](https://www.donmccurdy.com/2021/07/06/personal-updates-leaving-google/). Between this personal announcement and noticing his [developer "tip jar" on GitHub](https://github.com/sponsors/donmccurdy), I figured that he would be open to at least discuss the possibility of working together.\* I sent him an email. No reply. I followed up. No reply. The funny thing was that during this time where I couldn't get any response to my emails, Don was incredibly generous with answering lots of my programming questions that were related to the problems we were having with [polySpectra AR](https://ar.polyspectra.com). I quickly realized that if I was showing respect for his craft, showing I had done enough searching to attempt to solve the problem myself (or have our developers try to solve the problem themselves), and if I presented the question in the channels of Don's choice (GitHub & Discord at the time) --- then I was fairly likely to get Don to solve my problems within a day or two. Completely for free, on his own time, just for fun. I literally could not pay this expert. I could not get him to engage in a discussion about money. I could not buy his attention. But by engaging in a real dialog in an appropriate forum, I could see that Don was clearly enrolled in the idea of helping other people solve their problems with web AR. He was particularly helpful when I asked questions about how to use the amazing library that he wrote, [glTF Transform](https://gltf-transform.dev/). This makes sense, it is his project, he cares deeply about it, and he is enrolled in helping other people use it.\** Don is not an outlier. This is a very common pattern I have seen with the best in the world. If you offer money, you get no reply. If you engage in a genuine dialog, you can get the best advice, from the smartest people, for free. Money can't buy you enrollment. --- \* Re-reading this, I am laughing out loud at the way this phrase slipped in: "working together". But that's exactly what this is all about. What does it mean to "work together"? \** It looks like Don is now experimenting with a new way to monetize his expert attention: [glTF Transform Pro](https://store.donmccurdy.com/l/gltf-transform-pro). ### BREAKING NEWS: My Internet is Out - URL: https://raw.works/breaking-news-my-internet-is-out/ - Source: content/works/breakingnewsnointernet.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-26T09:05:15-07:00", "image": "/images/internetwarzone.jpeg", "lastmod": "2023-10-26T09:05:15-07:00", "publishdate": "2023-10-26T09:05:15-07:00", "tags": [], "title": "BREAKING NEWS: My Internet is Out" } ``` Content: ### A First-Person Account from the Frontlines The battle rages on here at my home front as I've just lost connection to the information superhighway. Absolute chaos has erupted - I've scrambled to deploy my personal hotspot reserves but even they can only offer up a trickle of bandwidth. Rations are being put in place - I've had to limit who else in the household can piggyback off my mobile connection at a given time. Streaming anything heavier than mere tweets appears to be out of the question for now. Moral is low amongst the ranks. We're all tired of staring at the dreaded buffering symbols that mock us endlessly. I've logged an urgent message with our ISP chatbot but so far their support has been insufficient. What began as chaos has descended into full blown crisis...particularly for the younger recruits who have never known life without a steady internet stream. The Millenial and Gen Z troops are flailing without the online world they're so accustomed to. Endless refreshes and buffering frustration has taken its toll on morale. Making matters worse, many have never had to endure these kinds of communications blackouts before. It's all they've ever known. Now they're scrambling to learn archaic skills like entertaining themselves without screens or getting homework done via mobile hotspots with data limitations. I'm fighting to keep a steady leadership presence but it's not easy wrangling troops so used to constant connectivity. Rations remain tight and options few. Unless the cavalry arrives soon with repairs, I fear a full meltdown may be imminent amongst the digitally dependent ranks. For now all I can do is pray the older generations can keep the younger ones from losing it completely until reinforcements liberate us from this internet Iron Dome. Wish us luck out here on the dark network front! Updates to follow if we can keep it together. Options are running thin fast. I'm seriously contemplating having to retreat to the office just to get some decent work done. Already non-essential and bandwidth-limiting programs, like VPN, have been cut off. {{< figure src="/images/internetwarzone.jpeg" title="A Brigade of Personal Hotspots to the Rescue" >}} I managed to scrape together just enough data to transmit this rough photo giving a glimpse of the tense atmosphere (seen above). Even falling back to being "that guy" who turns off his video during our next Zoom meeting is being quietly mulled as a potential last resort. As night falls with no repairs in sight, the fight to regain control of my internet access continues. I'll keep transmitting updates from the digital warzone as events allow, but for now all any of us can do is keep our fingers crossed that reinforcements arrive soon to liberate our stopped up information lines. Wish us luck! ### Poe Is My New Search Engine - URL: https://raw.works/poe-is-my-new-search-engine/ - Source: content/works/poeismynewsearchengine.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-25T13:28:28-07:00", "image": "/images/AI-inspector-gadget.jpeg", "lastmod": "2023-10-25T13:28:28-07:00", "publishdate": "2023-10-25T13:28:28-07:00", "tags": [], "title": "Poe Is My New Search Engine" } ``` Content: This post is about the attentional leverage of searching through AI, specifically with the [Poe Web Search](https://poe.com/Web-Search) bot. Sometimes the AI gives me the answer in the summary, sometimes it gives me the link where I need to dig deeper, sometimes it doesn’t find anything good. Regardless...there is no attentional rabbit hole. What are my figures of merit? - _Time to a good answer_ --- the time cost or opportunity cost. - _# of tabs opened to get to a good answer_ --- the cost of attention-switching, which I'm really trying to avoid. One of the traps of a traditional search engine is that it really feels like you’re doing important work. You are on a detective case. You are weeding out the nonsense. You are finding the hidden gems. If you truly enjoy searching and browsing for the sake of searching and browsing, then more power to you and please don’t let me ruin your pastime. But I ain’t got time for that ish. More importantly, I ain’t got attention for that ish. I want the answer now, even if it’s just to know that there isn’t a good answer on the web. If I can't have it now, then I want the answer with as little attention switching as possible. In my experience, [Poe Web Search](https://poe.com/Web-Search) is “smarter” than Bing. In other words - better results and no ads. What’s even cooler is that once I get the answer, I can quickly input that into one of many very useful Poe bots: [Claude 2](https://poe.com/Claude-2-100k), [GPT-4](https://poe.com/GPT-4), [Code Llama](https://poe.com/Code-Llama-34b), [6 Hats Helper](https://poe.com/6Hat-Helper), etc. Poe is kind of like Inspector Gadget - there is a chat bot for every job. I'm trying [not to make recommendations](/advice-under-uncertainty/) these days. Just sharing what works for me. {{< figure src="/images/AI-inspector-gadget.jpeg" title="AI Inspector Gadget" >}} ### pSai and Resina - URL: https://raw.works/psai-and-resina/ - Source: content/works/psaiandresina.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-24T18:12:08-07:00", "image": "/images/scifiaicuration.jpeg", "lastmod": "2023-10-24T18:12:08-07:00", "publishdate": "2023-10-24T18:12:08-07:00", "tags": [], "title": "pSai and Resina" } ``` Content: Today, we released two new AI chat bots that we have been building at polySpectra: [pSai](https://docs.polyspectra.com) & [Resina](https://resin3d.ai). [Here's the announcement post.](https://polyspectra.com/blog/two-new-ai-chat-bots-to-accelerate-your-resin-3d-printing-journey/) [pSai](https://docs.polyspectra.com) is trained on all of polySpectra's technical product information. We're now using it as an AI-augmented search function, for the entire [polySpectra.com](https://polyspectra.com) website. I wrote about my initial experience building pSai [here](https://raw.works/my-first-executive-ai-win/). [Resina](https://resin3d.ai) is a much more challenging project. Our goal is to train it on all of resin 3D printing. This will be pretty hard. Resina is accessible at [resin3D.ai](https://resin3d.ai). Only a couple hours after emailing our list, we already have a volume of un-answered (or poorly-answered) questions that would be pretty overwhelming for a lowly human to try to respond to on their own. I'm excited by the challenge to figure out a scalable system from the start. We're refining a setup where [Resina](https://resin3d.ai) learns a bit more every night. So, even if it can't answer a certain question today, the hope is that it'll know the answer by tomorrow. There are still a lot of humans-in-the-loop, but we're trying to make the system as automated as possible. (With humans as curators/editors of the AI's knowledge base.) We're aiming for something like this: {{< figure src="/images/scifiaicuration.jpeg" title="Rough Diagram of the Resina AI Workflow" >}} A few fun experiences from today: - We must be doing something right, because we already have a few really angry dialogs with people who appear to be major haters of polySpectra (and maybe AI too?) - I've been having a lot of fun with the combination of GitHub CoPilot and [Cursor.sh](https://cursor.sh). CoPilot quickly tries to autocomplete while Cursor can really do the full-context heavy lifting. These tools are designed for developers (and I am not a developer), but I'm finding them to be really useful for content creation. (See [yesterday's post:](/my-journey-with-cursor-a-new-wave-in-coding/) ) It's surprisingly good at coming up with new questions to ask Resina. - Very early exploration with the possibility of giving Resina a voice. [ElevenLabs](https://elevenlabs.io/) is super impressive. ### My Journey with Cursor: A New Wave in Coding - URL: https://raw.works/my-journey-with-cursor-a-new-wave-in-coding/ - Source: content/works/newpostaboutcursor.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-23T00:00:00Z", "image": "/images/cursor-demo-screenshot.png", "lastmod": "2023-10-23T00:00:00Z", "publishdate": "2023-10-23T00:00:00Z", "tags": [], "title": "My Journey with Cursor: A New Wave in Coding" } ``` Content: *Note: Watch how I wrote this (super cheesy) post in ~7 minutes with [Cursor](https://cursor.sh), via the new [Zoom Clips Beta Feature](https://us02web.zoom.us/clips/share/A2F3MSDq_1m5oEw_NM1OVzWmaQZdwVdlZW66923BeP086ksDQA)* {{< figure src="/images/cursor-demo-screenshot.png" title="Writing about Cursor with Cursor." >}} --- I've been exploring [Cursor](https://cursor.sh), an AI-first code editor that's making waves in the developer community. It's being used by engineers at big names like Shopify, Samsung, OpenAI, and Facebook, to name a few. But what's all the fuss about? Well, [Cursor](https://cursor.sh) is packed with features that are designed to make coding faster and more efficient. For starters, it lets you chat with your project. No more wasting time hunting for the right place to start a change or the correct method to call. It's like having a knowledgeable coding buddy right there with you. But that's not all. [Cursor](https://cursor.sh) also allows you to browse documentation directly and refer to code definitions and files. It's like having a coding reference library at your fingertips. One feature that really stands out is [Cursor's](https://cursor.sh) ability to make code changes. It's like having an AI assistant that can write low-level logic for you. Need to change an entire method or class? Just give it a prompt. Want to generate code from scratch? Just give it a simple instruction. It's a game-changer. And let's not forget about debugging. [Cursor](https://cursor.sh) can scan your code for bugs and help you fix them quickly. It's like having a personal bug detective that automatically investigates linter errors and stack traces to figure out the root cause of your bug. I've been using [Cursor](https://cursor.sh) every single day (since yesterday 😜), and it's become an integral part of my coding routine (today). It's not just the advanced features that make it stand out, but also the ease of use and the efficiency it brings to my workflow. It's more than just an editor, it's a tool that truly understands the needs of a developer. So, is [Cursor](https://cursor.sh) worth the hype? Well, it's loved by developers all over the world and has helped tens of thousands of them be more productive. It's not just an editor; it's a game-changer in the world of coding. And I'm excited to see where it goes from here. ### Finding the Perfect Voice Transcription Tool - URL: https://raw.works/finding-the-perfect-voice-transcription-tool/ - Source: content/works/voice2text.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-22T00:00:00Z", "image": "/images/abstract-voice-to-text.jpeg", "lastmod": "2023-12-08T00:00:00Z", "modified": "2023-12-08T00:00:00Z", "publishdate": "2023-10-22T00:00:00Z", "tags": [], "title": "Finding the Perfect Voice Transcription Tool" } ``` Content: For the last four or five months, I have been on the hunt for the best voice to text transcription tool. This is a really important part of my workflow and helps me to be able to work while I am walking or driving. (I use the [Yealink BH71 Pro headset](https://amzn.to/3QaQ3ii), which has amazing noise cancelling (for the microphone) and is still small enough to wear while active.) In my quest for the perfect transcription tool, I've tried and tested numerous options. Here, I present a detailed comparison of three of the many tools I've explored: [Otter](https://otter.ai/referrals/QCJ7HSUD), [WhisperBoard](https://apps.apple.com/us/app/whisperboard/id1661442906), and [HappyScribe](https://www.happyscribe.com/): **[Otter](https://otter.ai/referrals/QCJ7HSUD):** - Has a nice mobile app - Provides rough transcription on the fly - Refines transcription when you have a connection - Free plan is fairly generous, you can use it to get in the habit. - Cheapest subscription is $16.99/month **[WhisperBoard](https://apps.apple.com/us/app/whisperboard/id1661442906):** - Completely free and open source - Runs locally on your phone for maximum privacy - Can choose from various accuracy levels, but the bigger/more accurate models are slow - Sometimes gets stuck repeating phrases - Not updated frequently, seems to be a hobby project **[HappyScribe](https://www.happyscribe.com/):** - Not optimized for mobile - Highest accuracy transcription I've seen - Can add custom vocabulary to improve accuracy - Took longest to find (no mobile app) - User interface designed for desktop - Only 10 minutes free transcription for testing - Cheapest subscription is $17/month In summary, [Otter](https://otter.ai/referrals/QCJ7HSUD) is best for quick rough transcription on mobile, [WhisperBoard](https://apps.apple.com/us/app/whisperboard/id1661442906) is best for full privacy but you get what you pay for it, and [HappyScribe](https://www.happyscribe.com/) has the highest accuracy but is designed for desktop. The price difference between [HappyScribe](https://www.happyscribe.com/) and [Otter](https://otter.ai/referrals/QCJ7HSUD) is only $0.01/month. They each have tradeoffs between accuracy, mobility, privacy and cost. | Feature | [Otter](https://otter.ai/referrals/QCJ7HSUD) | [WhisperBoard](https://apps.apple.com/us/app/whisperboard/id1661442906) | [HappyScribe](https://www.happyscribe.com/) | |-|-|-|-| | Mobile App | ✅ Nice native app | ✅ Runs on phone | ❌ Only browser | | Transcription Quality | ❌ Meh | ⚖️ Varies by model | ✅ Highest accuracy | | Privacy | ❌ Uses cloud | ✅ Runs locally | ❌ Uses cloud | | Cost | ✅ Free & $16.99/mo | ✅ Completely free | ❌ $17/mo minimum | | Customization | ❌ None | ❌ None | ✅ Custom vocab | | Testing Options | ✅ Free is generous | ✅ Completely free | ❌ Only 10 mins free | Right now I'm still in the final phases of what will likely be a decision to go all-in on [HappyScribe](https://www.happyscribe.com/). I'm recording in [Otter](https://otter.ai/referrals/QCJ7HSUD), then exporting the audio, re-importing to [HappyScribe](https://www.happyscribe.com/), and comparing the difference. If I were working at [Otter](https://otter.ai/referrals/QCJ7HSUD), I would put some focus and attention on transcription accuracy. If I were working at [HappyScribe](https://www.happyscribe.com/), I would build a mobile app. Honorable mention goes to the built-in iOS voice-to-text, which was used to dictate much of this post. (And in case you are curious, here was my process of dictating to Claude-instant-100k, to get most of this post written without using my hands => [https://poe.com/rawworks/1512928000195140](https://poe.com/rawworks/1512928000195140) ) --- Update 12/8/23: I switched to [MacWhisper](https://goodsnooze.gumroad.com/l/macwhisper) => [AI Transcription Without a Subscription](/ai-transcription-without-a-subscription) ### It's All in the Hips - URL: https://raw.works/its-all-in-the-hips/ - Source: content/works/itsallinthehips.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-21T00:00:00Z", "image": "/images/allinthehips.png", "lastmod": "2023-10-21T00:00:00Z", "publishdate": "2023-10-21T00:00:00Z", "tags": [], "title": "It's All in the Hips" } ``` Content: This is what I've been up to today: {{< youtube id="twrBEonqt-Q" >}} For this type of "awareness through movement", I would consider [low-dose ketamine](/a-joyous-equinox/) to be a performance-enhancing drug. (& of course I found [Alfons](https://www.alfonsgrabher.com/i-asked-chatgpt-to-teach-me-a-feldenkrais-lesson/?ref=rawworks) by asking Poe Web Search for "Feldenkrais AI".) ### GPT-4V(ision) Handwriting OCR - URL: https://raw.works/gpt-4vision-handwriting-ocr/ - Source: content/works/gpt4v-handwritingOCR.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-20T00:00:00Z", "image": "/images/gpt4v-handwritingocr.png", "lastmod": "2023-10-20T00:00:00Z", "publishdate": "2023-10-20T00:00:00Z", "tags": [], "title": "GPT-4V(ision) Handwriting OCR" } ``` Content: Last month I was whining about "handwriting OCR", a technology that has mysteriously and simultaneously existed in a quantum superposition between |*in use every day at the USPS*⟩ and |*doesn't actually work*⟩ since at least 1998. (See [Why Does Handwriting OCR Suck in 2023?](/why-does-handwriting-ocr-suck-in-2023/)) GPT-4 with Vision (GPT-4V) is now available to ChatGPT Plus users (somehow Microsoft didn't release this one first). I figured I'd take a break from my ranting and put it to the test: {{< figure src="/images/gpt4v-handwritingocr.png" title="GPT-4V transcribes my notebook into Markdown." >}} It didn't nail the indentation of the bullet points, but I think this is getting pretty close to useful. If anyone can think of a way to fine-tune GPT-4V (I think we may need to wait for the API) - please let me know. Maybe I'll finally be able to digitize my handwritten notebooks this decade. Shout out to "All About AI" for showing the example that inspired me to revisit this. At [00:32](https://www.youtube.com/watch?v=Yq2VOWDFpNA&t=32s) - he draws an outline of a program in his notebook and asks GPT4 to write the corresponding code. Full video below: {{< youtube Yq2VOWDFpNA >}} ### Using AI at Every Step of the Customer Journey - URL: https://raw.works/using-ai-at-every-step-of-the-customer-journey/ - Source: content/works/aiforcustomerjourney.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-19T00:00:00Z", "image": "/images/AI-customer-journey.png", "lastmod": "2023-10-19T00:00:00Z", "publishdate": "2023-10-19T00:00:00Z", "tags": [], "title": "Using AI at Every Step of the Customer Journey" } ``` Content: In the same way that every company needed to become an "Internet company”, and then have a social media presence --- now every company is going to need to become an "AI company". My first professional "AI win" was [building a customer-support chatbot](/my-first-executive-ai-win/). But how about other stages of the customer journey? This post is me "learning out loud", to show y'all what I've been up to. I'm going to use the simple framework of Attract > Convert > Close > Delight. ## Attract - Social Media images, [DALL·E 3 Can Almost Spell](/dalle-3-can-almost-spell/) {{< figure src="/images/AI-customer-journey.png" title="AI can spell AI, that's a start." >}} - Expertise in the market niche: "Meet Resina - Your Resin 3D Printing AI Agent" https://resin3d.ai/ ## Convert - For polySpectra's engineering audience, AI answering questions about the documentation is arguably an important part of conversion, closing, and delighting customers. Here's our support bot in action: [https://docs.polyspectra.com/](https://docs.polyspectra.com/) - Similarly, one of my goals with [Resina](https://resin3d.ai/) is to freely educate people as much as possible about resin 3D printing. [polySpectra](https://polyspectra.com) was founded to solve the founding technical debt of stereolithography 3D printing (shitty materials) - so the faster people learn about the "gotchas" of resin printing, the faster they'll realize that they need materials they can actually trust. ## Close - This stage warrants extra precaution. Depending on the type of interaction that you are having, AI could be a surefire way to destroy any trust that you've built with the prospect. I'm not sure I have good advice for a true sales close, other than perhaps using AI to help word emails. I think the best approach is using AI to be so helpful and so informative (in the other stages) that the customer closes themselves. ## Delight - Speed. I'm hooked on same-day shipping. I can't wait 5 seconds for ChatGPT to respond, I need Claude-Instant instead. AI can help delight customers by accelerating customer engagement and customer support 100-1000x. - "Mass customization". This is a term thrown around a lot in 3D printing. LLMs can be fine-tuned so quickly now (see [How To Train ChatGPT On A Book In 5 Minutes](/how-to-train-chatgpt-on-a-book-in-5-minutes/)) that you could imagine customizing the AI to be specific to an individual customer. I'm excited to explore this idea further. - Make it fun. As we start to build more and more AI tools at [polySpectra](https://polyspectra.com), I'm trying to have fun with it. I'm also trying to build tools that I would legitimately want to use. ### Searching for Super-Discernment - URL: https://raw.works/searching-for-super-discernment/ - Source: content/works/superdiscernment.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-18T00:00:00Z", "image": "/images/headorbit.jpeg", "lastmod": "2023-10-18T00:00:00Z", "publishdate": "2023-10-18T00:00:00Z", "tags": [], "title": "Searching for Super-Discernment" } ``` Content: Over the years, I have had lot of "good" ideas that just clearly weren’t worth my time. Now that AI can accelerate some of these ideas by 100x or maybe even 1000x - does that make them worth pursuing? Certainly for some of them. The super-doing capabilities of AI require super-discernment capabilities for humans. The possibilities are completely overwhelming to me, every day I find myself paralyzed by the sheer scale of potential permutations. I feel something like this: ![paralyzed by potential permutations](/images/headorbit.jpeg) ### Remembering Roland Griffiths - URL: https://raw.works/remembering-roland-griffiths/ - Source: content/works/rememberingroland.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-17T00:00:00Z", "image": "/images/rolandandraymond.JPG", "lastmod": "2023-10-17T00:00:00Z", "publishdate": "2023-10-17T00:00:00Z", "tags": [], "title": "Remembering Roland Griffiths" } ``` Content: The first time that I met Roland Griffiths, he was setting up someone else's tent at Burning Man. This small act of generosity - setting up someone else's tent - might not sound like a big deal, but it turned out to be a really big deal for me. (I happen to now be married to that someone.) Roland was a truly inspiring friend and mentor to me. He had incredible discipline in all areas of his life - his work ethic, his self-care routines, his meditation practice. From the moment I met him, I could tell immediately that he had a very clear sense of purpose, a real mission, perhaps best conveyed by his favorite question: *"Are you aware that you are aware?"* You can likely sense Roland's child-like curiosity in this question. His curiosity wasn't just about ideas, Roland was simply one of the most loving and compassionate people that I have ever met. I am incredibly grateful for the time I had with him over the last decade or so. {{< figure src="/images/rolandandraymond.JPG" title="Roland Griffiths teaching me how to meditate, as a wedding gift." >}} Professionally, Roland was a rockstar. He was one of the world's foremost experts in seemingly disparate subjects: caffeine, mystical experiences, and psilocybin. In ["How to Change Your Mind"](https://amzn.to/3QkqtJ3), Roland is described as "the investigator beyond reproach". Roland was among the last of a generation of researchers to use self-experimentation as a mode of scientific investigation. Once regarded as the gold standard, this is now considered to be too conflicted and biased to be used in modern research (which currently favors double-blind controlled studies). For example, one of Roland's academic contributions that involved self-experimentation was towards quantifying the minimum detectable dose of caffeine. (This was at a time when caffeine was not considered to be addictive, and not treated as "a drug".) Roland was truly at the cutting edge of research in consciousness. He was incredibly curious about probing the line between science and spirituality. With the utmost respect for my dear friend, many of the questions he was seeking to explore in his research were beyond science. Roland was not afraid to look over that edge. What do I mean by beyond science? To me, science stops when we can't propose a means for others to reproduce the experiment. So many of the questions about consciousness (not just Roland's) are fundamentally irreproducible. It's not that they aren't interesting questions - it's that these are philosophical questions, these are spiritual questions, these are mystical questions. In my view they cannot be answered with the tools of physical science. (But hey, I'm just a lowly chemist.) I am forever grateful to Roland for these debates. I don't know exactly where to draw the line between science and spirituality, but my conversations with him helped me to start sketching a rough topology. Without getting too heady, let me explain what I mean with a short example: {{< blockquote >}} *Roland Griffiths taught Michael Phelps how to swim.* {{< /blockquote >}} We can test this statement with scientific investigation. We might never be able to prove without a doubt that Roland did or did not teach Michael how to swim (it takes a village), but if we asked enough people at the pool we could probably get a good scientific sense of the degree to which Roland influenced Michael's swimming abilities. {{< blockquote >}} *Roland Griffiths and Michael Phelps have the same conscious experience while swimming.* {{< /blockquote >}} This one gets tricky. Some people would argue that we just don't have sharp enough scientific tools to probe that question. To me, this question is beyond science. From a scientific perspective - it is unknowable, even while from a psychological perspective Roland and Michael both intimately know the experience of swimming. My view is that no matter how many words, sounds, EEGs, fMRIs, or nerve clamps that we use - science alone can't compare their direct conscious experience. (And the beauty of it is that Michael turned out to be a decent swimmer, even though we'll never be sure if Roland's experience of swimming even remotely resembled Michael's.) That's enough philosophy of mind for today (see the footnotes for some of my favorite books on this topic). Let's get back to remembering my buddy Roland. Roland was unbelievably upbeat and cheerful after receiving the news of his terminal cancer diagnosis. *"What a gift"*, he would say. *"What a tremendous opportunity to really think about my priorities."* While I can't imagine myself responding to a similar situation with even a fraction of his equanimity - I don't doubt that he meant it. I certainly don't have the words for it now, but I feel that I have learned a tremendous amount from Roland recently, observing how he has courageously and compassionately navigated his final months on earth. Towards the end of his life, Roland was very curious about the possibility of communicating after death. Since I heard the news of his passing yesterday, I've been quietly investigating this possibility. I can't prove it scientifically, Roland, but I feel your presence. I feel your love. Thank you. --- Footnotes: - [Roland Griffiths Memorial Fund](https://griffithsfund.org/) - ["How to Change Your Mind" - Michael Pollan](https://amzn.to/3QkqtJ3) - ["Galileo's Error" - Philip Goff](https://amzn.to/3ZYuGVW) (special thanks to Dana Sawyer for gifting me this book) - ["The Flip" - Jeff Kripal](https://amzn.to/3Fjc0Xs) (also via Dana) ### Leveraging AI's Alexithymia to Overcome Emotional Friction at Work - URL: https://raw.works/leveraging-ais-alexithymia-to-overcome-emotional-friction-at-work/ - Source: content/works/ai_alexithymia.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-16T00:00:00Z", "lastmod": "2023-10-16T00:00:00Z", "publishdate": "2023-10-16T00:00:00Z", "tags": [], "title": "Leveraging AI's Alexithymia to Overcome Emotional Friction at Work" } ``` Content: The emotional component is one of the toughest parts of the job. Whether it's communicating with someone you recently had a conflict with, following up on a tedious but important task, or just generally pushing through moments of resistance or procrastination - our emotions don't always make things easy. What if we could leverage the natural "[alexithymia](https://en.wikipedia.org/wiki/Alexithymia)" of AI systems to our advantage in overcoming emotional friction? As AI's don't appear to have subjective experiences (yet), they are impervious to emotions like frustration, resentment, burnout or reluctance that humans often face (for now). Imagine there's someone you need to follow up with, but just the thought of directly interacting with them starts to raise your blood pressure. Use AI to draft the message. This would allow you to communicate pragmatically without letting any lingering interpersonal tensions interfere. The AI's objective, emotion-free communication style would sidestep your own anxious reactions. So rather than dwelling on past issues or stressing over an awkward encounter, you focus solely on conveying the necessary information. The AI acts as a buffer, providing a bit of distance from an emotionally fraught situation. This helps you still get your point across and move things forward, while saving your emotional energy for interactions that truly matter. Similarly, if there's a task I've been resisting or procrastinating due to fatigue or lack of motivation, enlisting an AI assistant can help separate the emotional component from the logical "next steps". This allows me to leverage a methodical, emotion-free collaborator to help move projects forward despite any personal reservations. Showing up is 90% of the work, and LLMs "show up" in milliseconds, with minimal pay. By reducing the emotional friction of the work - they can help me show up even when I don't really feel like it. Some prompts to experiment with: - Have an AI assistant draft a difficult email, message, or presentation you've been avoiding. Remove your emotional reaction from the content. - Enlist an AI to break a complex task down into bite-sized, manageable steps so it feels less overwhelming. - Give an AI routine administrative tasks you've come to dislike through repetition so you can focus on more engaging work. - Utilize AI databases, research tools, or information retrieval to find answers independently rather than expose yourself to the emotional slot-machine of the internet - [Develop an AI chatbot to field basic user questions and issues](my-first-executive-ai-win/) as the first point of contact, reserving your expertise for more nuanced inquiries. - Use an AI to guide you (and your team) through a perspective-switching exercise (for example, I've been using my ["6 Hats Helper" bot](https://poe.com/6Hat-Helper) almost every day). With great power comes with great responsibility. These tools could be used to help you become even more emotionally avoidant, or they could be used to reserve your human energy and attention for the truly important and highly emotional work. ### Ongoing Moving Sale - Text For Info - URL: https://raw.works/ongoing-moving-sale-text-for-info/ - Source: content/works/movingsaletextforinfo.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-15T00:00:00Z", "image": "images/ongoingmovingsale.jpg", "lastmod": "2023-10-15T00:00:00Z", "publishdate": "2023-10-15T00:00:00Z", "tags": [], "title": "Ongoing Moving Sale - Text For Info" } ``` Content: ![Ongoing Moving Sale](images/ongoingmovingsale.jpg) In a small village surrounded by terraced rice fields, there stood a run-down storehouse. Out front lay a lone piece of weathered plywood, spray painted with odd markings that had faded with time: “ONGOING MOVING SALE! TEXT FOR INFO”. A phone number was listed below, but no one had ever paid it any mind. The villagers assumed the building must contain only trash, since its owner had clearly left in a hurry, judging by the messy sign. Over the years, the storehouse had fallen into total ruin and disrepair. One sunny day, as the village children played, curious young Tomas spied the mysterious sign and could not resist investigating further. Peeling back flakes of faded paint, he decoded the message and decided to text the number. There was no reply. Tomas could not control himself. The next day, he decided to go in anyway. When Tomas pushed open the creaking door, layers of dust stirred... What wonders were then revealed to Tomas's keen eyes! On wooden shelves lining the walls from floor to ceiling were stacked objects untouched for who knows how long. Intricate sculptures from ancient civilizations were piled high, their artistic details still visible under years of dust. Figures of elephants, horses and dancers from India appeared as if in meditation. Weaving amongst them were relics from every corner of the universe. From distant galaxies, celestial maps charted nebulae and star systems in vibrant cosmic oils. Alien artifacts of long-lost civilizations spoke to lives pondering the same mysteries under alien skies. Sculptures hewn from asteroids captured the shifting dances of stardust with microscopic perfection. In dark corners, musical instruments of all shapes and sizes from a thousand worlds rested in silent harmony. Zithers, flutes and drums from lands afar awaited the songs they held within. Rolled parchments and faded tomes filled with the wisdom of sages from days long forgotten. Secrets of medicine, philosophy and history that could lift entire communities from suffering were here in abundance. From Ancient Egypt emerged papyrus scrolls preserving the wisdom of Imhotep, yielding knowledge to lift humankind from sickness. Statuettes and funerary masks channeled the poise of the pharaohs, as membranes within guided souls to sacrifice all for their people. Opal scarabs and gems engraved with spells escorted souls to paradise realms described only in poems. The glories of Rome were not absent. Marble friezes recounted triumphs through disciplined legions and advances in architecture that stood the test of time. From distant Guangdong came intricate bronzewares and jades inscribed with early dialects, as were silken banners from Chang'an imprinted with astrological charts. Scrolls of rice paper bore poetic odes and wisdom that fed countless souls through dynasties. From the deserts of a thousand worlds surrendered their secrets. Scrolls penned in luminescent sands illuminated the insights of hermit monks who dwelled alone with dunes for decades. Elixirs distilled from meteorites held properties to cure afflictions unnamed. Even the roots of World Trees bowed under the abundance. Folios bound in barkskin preserved the visions of prophets who communed for centuries in leafy groves. Carvings in flowing mycelium recorded the epics of subterranean elders conversing telepathically in living rootworks. From dimensions unseen, objects of pure imagination took shape. Sculptures of mercy and grace sculpted from liquid dreams. Harps whose strings were spun from strands of thought provoking melodies never before conceived. Prisms refracting the nuances of concepts beyond words into rainbows of meaning. As Tomas explored further into the depths, the piles of marvels seemed to grow inexhaustibly. He stumbled upon a veritable library of Babylonia - clay tablets bearing epics, etchings of constellations, and arithmetic used to chart the heavens. Inkwashed scrolls by masters from Chang'an imparted the subtle and profound. Lost amid wonders unimaginable, Tomas forgot all about the passage of time in his awe and delight. Lost amid wonders unimaginable, Tomas forgot all about the passage of time in his awe and delight. Truly, this place held riches utterly without measure or bound, enough to uplift all beings throughout the cosmos for eras untold. ### Generous Residue - URL: https://raw.works/generous-residue/ - Source: content/works/GenerousResidue.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-14T00:00:00Z", "lastmod": "2023-10-14T00:00:00Z", "publishdate": "2023-10-14T00:00:00Z", "tags": [], "title": "Generous Residue" } ``` Content: I was given a gift and en route to enjoyment The sticky residue of the price tag Stuck to my fingers Pulpy bits in my nail beds I never could remove No soap nor scour strong enough To undo the glue My fingerprints changed Possibly my DNA Incredible shame Unforgettable blame Enraged and enflamed All in the name of erasing the price of the gift ### Books Are So 20th Century - URL: https://raw.works/books-are-so-20th-century/ - Source: content/works/booksareso20thcentury.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-13T00:00:00Z", "lastmod": "2023-10-13T00:00:00Z", "publishdate": "2023-10-13T00:00:00Z", "tags": [], "title": "Books Are So 20th Century" } ``` Content: This is the last moment in history where the act of writing a book will be considered impressive. Just like Uber was “everyone’s private driver”, LLMs are now “everyone’s private ghostwriter”. Both Bill Gates & HH the Dalai Lama have ghostwriters - why shouldn’t you? --- Here are just a few pieces of evidence: - [Amazon restricts authors from self-publishing more than three books a day after AI concerns](https://www.theguardian.com/books/2023/sep/20/amazon-restricts-authors-from-self-publishing-more-than-three-books-a-day-after-ai-concerns) - [How To Train ChatGPT On A Book In 5 Minutes](/how-to-train-chatgpt-on-a-book-in-5-minutes/) - [Do AI detectors work? In short, no...](https://help.openai.com/en/articles/8313351-how-can-educators-respond-to-students-presenting-ai-generated-content-as-their-own) ### Squeezed By The Universe - URL: https://raw.works/squeezed-by-the-universe/ - Source: content/works/squeezedbytheuniverse.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-12T00:00:00Z", "image": "/images/squeezedbytheuniverse.jpeg", "lastmod": "2023-10-12T00:00:00Z", "publishdate": "2023-10-12T00:00:00Z", "tags": [], "title": "Squeezed By The Universe" } ``` Content: Being squeezed by the universe is usually quite uncomfortable. A loving kick in the pants from a higher dimension. A (re)birth canal that will change the shape of your head. An invitation not to the party you dreamed of, but to the path you belong to. ![Squeezed by the universe](/images/squeezedbytheuniverse.jpeg) ### 6 Hats Helper: Your New Thinking Buddy - URL: https://raw.works/6-hats-helper-your-new-thinking-buddy/ - Source: content/works/6hatshelper.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-11T00:00:00Z", "image": "/images/6Hats.jpeg", "lastmod": "2023-10-11T00:00:00Z", "publishdate": "2023-10-11T00:00:00Z", "tags": [ "1st Time", "_test" ], "title": "6 Hats Helper: Your New Thinking Buddy" } ``` Content: Ever had one of those days where every decision feels like a brain-bender? Well, we've been there and we got you covered. Meet [6 Hats Helper](https://poe.com/6Hat-Helper), my latest AI chatbot. But hold on, it's no ordinary chatbot. It's like your personal decision-making sidekick, inspired by the genius of [Edward de Bono's Six Thinking Hats](https://amzn.to/3ZRGbhP) technique. Picture this. The [6 Hats Helper](https://poe.com/6Hat-Helper) is like your own cognitive jiu-jitzu sensei, wearing Blue, White, Red, Black, Yellow, and Green hats, each one representing a different style of thinking. It's as if you've got a team of top-notch thinkers in your head, each bringing their unique perspective to the table, helping you make decisions that cover all bases. Whether you're wrestling with a mind-boggling problem or trying to strategize your next big move, the [6 Hats Helper](https://poe.com/6Hat-Helper) steps in to guide your thought process. It's like having a structured, multi-angle approach to handling any challenge, right in your pocket. With the [6 Hats Helper](https://poe.com/6Hat-Helper), decision-making isn't just about 'eeny, meeny, miny, moe' anymore. It's a thorough expedition into the world of thoughts. It's like having an advisory board in your brain, each member giving you a different viewpoint, making sure you've got all your angles covered. Building the [6 Hats Helper](https://poe.com/6Hat-Helper) was an absolute blast. Thanks to Poe and Claude-2, creating this bot was way easier than I anticipated. (This is my first Poe bot.) Are you ready to shake up your decision-making game? Put [6 Hats Helper](https://poe.com/6Hat-Helper) to the test and tell me what you think. ![6 Hats Helper](/images/6Hats.jpeg) ### How To Train ChatGPT On A Book In 5 Minutes - URL: https://raw.works/how-to-train-chatgpt-on-a-book-in-5-minutes/ - Source: content/works/howtotrainchatgptonabookin5minutes/index.md Front matter: ```json { "date": "2023-10-10T01:00:00-07:00", "lastmod": "2023-10-10T01:00:00-07:00", "publishdate": "2023-10-10T01:00:00-07:00", "title": "How To Train ChatGPT On A Book In 5 Minutes" } ``` Content: - Step 0: Get your book as a .PDF (As an example, I chose "The 4-Hour Work Week", which is [available as a PDF for free from the author](https://tim.blog/wp-content/uploads/2014/10/the-4-hour-workweek-expanded-and-updated-by-timothy-ferriss.pdf).) - Step 1: Sign up for SuperAgent: https://beta.superagent.sh and connect your OpenAI API key (by pasting it [here](https://beta.superagent.sh/llms) after clicking "configure"). - Step 2: Go to ["Datasources"](https://beta.superagent.sh/datasources). Click "New Data Source". Click "Local files". Pick the file, give it a name and description. ![New Data Source for SuperAgent](/how-to-train-chatgpt-on-a-book-in-5-minutes/newdatasource-superagent.png) - Step 3: (While your data source is processing...) go to ["Agents"](https://beta.superagent.sh/agents). Click "New Agent". Under Datasources, select the source. ![Select the Data Source](/how-to-train-chatgpt-on-a-book-in-5-minutes/selectdatasource.png) - Step 4: Give it a prompt. I chose this: {{< blockquote >}} _You are an helpful AI Assistant. You are an expert in productivity. You are an expert in startup business systems. When answering a question, first see if the answer is available in The Four-Hour Work Week book data._ {{< /blockquote >}} - Step 5: Choose the model and connect the datasource. Click "Update Agent" to save. Your agent setup should look something like this: ![Agent Setup](/how-to-train-chatgpt-on-a-book-in-5-minutes/agentsetup.png) - Step 6: Go back to ["Datasources"](https://beta.superagent.sh/datasources) to confirm that the datasource is done processing. It should say "Ready". ![Datasource Done Processing](/how-to-train-chatgpt-on-a-book-in-5-minutes/4HWWdatasource.png) - Step 7: Go back to your agent and test it out! Here was my test question: {{< blockquote >}} help me with my "dreamlining" excercise. to confirm your understanding, summarize the process of "dreamlining" from the book. {{< /blockquote >}} ...and the response: {{< blockquote >}} "Dreamlining" is a concept introduced by Tim Ferriss in his book "[The 4-Hour Work Week](https://amzn.to/3FcOr2z)". It's a process that involves defining your goals and dreams in concrete terms and then breaking them down into actionable steps. Here's a summary of the process: Create Two Timelines: Create two timelines - one for six months and one for twelve months. For each timeline, list up to five things you dream of having (including, but not limited to, material wants: house, car, clothing, etc.), being (be a great cook, be fluent in Chinese, etc.), and doing (visiting Thailand, tracing your roots overseas, racing ostriches, etc.) in that order. Determine the Cost: Next, you need to determine the cost of these dreams to make them a reality. This includes the cost of living your dream and any expenses that might arise in the pursuit of it. Define Action Steps: Break down the steps you need to take to achieve each dream. Start with the steps you can take today. Take the First Step: Now that you have your steps defined, take the first step immediately. This creates momentum and makes it more likely that you'll follow through with the rest of your plan. Remember, the goal of dreamlining is to make your dreams achievable by breaking them down into manageable, actionable steps. It's about turning the abstract into the concrete and making your dreams a reality. {{< /blockquote >}} It took me 9 minutes to figure out how to do this (including finding/downloading/uploading the 4HWW .pdf). (I gave myself a time limit of 10 minutes). If you follow these instructions (or ask your AI agent to follow them for you), I bet you can do it in 5 minutes. ### Crickets - URL: https://raw.works/crickets/ - Source: content/works/crickets.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-09T00:00:00Z", "image": "/images/tough-crowd.png", "lastmod": "2023-10-09T00:00:00Z", "publishdate": "2023-10-09T00:00:00Z", "tags": [], "title": "Crickets" } ``` Content: {{< center >}} A bold idea born, Shared with excitement, Met with tranquility, Crickets teach silence. {{< /center >}} ![Tough Crowd](/images/tough-crowd.png) ### (Some) Questions For a New Project - URL: https://raw.works/some-questions-for-a-new-project/ - Source: content/works/somequestionsforanewproject.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-08T00:00:00Z", "lastmod": "2023-10-08T00:00:00Z", "publishdate": "2023-10-08T00:00:00Z", "tags": [], "title": "(Some) Questions For a New Project" } ``` Content: ## Project Objectives and Scope - What are the goals and objectives of this project? - How does this project bring value to customers/society? - How will the capabilities created by this project support the organization's strategic goals? - What is the scope and deliverables? - If there are any non-negotiable constraints (e.g., regulatory, time, budget), what are they? ## Stakeholders - Who are the key stakeholders? - What are their expectations and influence? - How will stakeholder feedback be incorporated throughout the project? - How will we continue to sync with them? ## Team - Who is on the team and what are their roles? - Is the project aligned with team values and passions? - How will decisions be made as a team? - How will we handle skill gaps within the team, if any? - How will we handle conflicts within the team? - How will we celebrate wins and milestones together? ## Communication - How often will we meet for updates? - What communication tools or platforms will we use? - What anonymous channels can we use for feedback? - How will we document key decisions and who will be responsible for this? - How will we continue to communicate with stakeholders? ## Timelines and Milestones - What is the timeline and milestones? - Are there any schedule dependencies? - How flexible are the milestones and deadlines? Can they be adjusted based on project progress or unforeseen obstacles? - How can we build in feedback loops and iterations? ## Risks and Obstacles - What risks and challenges may arise? - What pitfalls have we faced before that we can now avoid? - How will we identify and monitor emerging risks as the project progresses? - Do we have backup plans and contingencies? - What could potentially derail this project? ## Resources - What resources are available? - What is our plan for resource allocation and management throughout the project? - Are any key resources missing? ## Sustainability - How can this project be sustained long-term? - How will the results of this project be maintained or enhanced after project completion? ## Project Management - What methodology will be used? - How will we track progress and reassess? - How will we manage scope creep, if it occurs? - How will we handle change requests? ## Success Criteria - How will we define and measure success? - What are the key performance indicators (KPIs) for this project? - What would it feel like to do a good job? - When and how will we reevaluate our approach? - What lessons learned from past projects can we apply to this one to increase its success? ### Purple Impressions - URL: https://raw.works/purple-impressions/ - Source: content/works/purpleimpressions.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-07T00:00:00Z", "lastmod": "2023-10-07T00:00:00Z", "publishdate": "2023-10-07T00:00:00Z", "tags": [], "title": "Purple Impressions" } ``` Content: I just finished my first week of [Purple Space](https://www.purple.space/), "a community of practice for creators, entrepreneurs, and line managers" founded by Seth Godin. Like any new community, things got off to a somewhat awkward start...feeling a bit disoriented and sheepish. Who is who? Where am I supposed to sit? What is normal here? How am I expected to engage? Do I want to be here? Today was a big inflection point. Seth organized a "Round Robin Chat" this morning, which consisted of about 50 people who were split into pairs for 5 minutes. There were 5 rounds, interspersed with a bit of commentary from Seth. The whole thing lasted less than an hour. I'm feeling inspired. It's a really simple container which can be easily implemented by Zoom. I found this "speed dating" format to be really energizing. In 5 minutes you almost always end up getting cut-off mid-sentence when the breakout room ends. This might sound frustrating but it's super fun. There's a cliff-hanger. It's hard to get bored in 5 minutes. You're left curious about what could have been if the interaction were longer. Afterwards, I couldn't help but think of how different the round robin "set and setting" was than most professional gatherings I've been to. The juxtaposition that immediately came to mind was a community/conference/company "happy hour" - there might be some conversations you're really trying to get out of, someone you're anxious/excited about approaching, someone you really want to avoid, that new person who joins in late and now the whole story has to be told again, someone buys you a beer and now you feel like you have to talk to them, someone offers to buy you a beer and now you feel like you have to drink, and on and on... The Zoom round robin is less organic, more mechanical. There isn't the opportunity to go very deep. But that's what's fun about it. Who goes first? How do I present myself in just a minute? How can I help someone in just a couple of minutes? Who wouldn't want to learn about the intricacies of alpaca fur for five minutes? Aside from just being a great way to meet five new people in the online community, I left feeling inspired to create community. I couldn't help but recall of the different types of events I've hosted in the past. Remembering that skill, that possibility, that muscle. ### When The Magic Is Gone - URL: https://raw.works/when-the-magic-is-gone/ - Source: content/works/whenthemagicisgone.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-06T00:00:00Z", "image": "/images/youngwizard.jpg", "lastmod": "2023-10-06T00:00:00Z", "publishdate": "2023-10-06T00:00:00Z", "tags": [], "title": "When The Magic Is Gone" } ``` Content: When the magic is gone... walk on the other side of the street do a rain dance hire a magician seek wise counsel draw a box around the sacred take the sacrament put down your weapons surrender your dictionary get very still observe. ### Indra's Web: Part 1 - URL: https://raw.works/indras-web-part-1/ - Source: content/works/indraswebpart1.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-05T00:00:00Z", "featuredimage": "/images/indrasweb1.jpeg", "image": "/images/indrasweb1.jpeg", "lastmod": "2023-10-05T00:00:00Z", "publishdate": "2023-10-05T00:00:00Z", "tags": [], "title": "Indra's Web: Part 1" } ``` Content: Indra's Web, or Indra's Net, is a cosmic net laced with jewels that hang over the palace of Indra, the king of gods in Hindu mythology. The net is infinite, and at each of its junction points there is a jewel that infinitely reflects the others. This is said to symbolize the infinite interconnectedness of all phenomena. Just as each jewel contains the reflections of all the others, each phenomenon in the universe contains within it all other phenomena. Nothing exists independently, but rather everything arises dependent on causes and conditions. What got me thinking about Indra's Net was the idea of building my own internal wiki for different things that I'm referencing often. For example if I'm referencing a particular book regularly, instead of just having the link/title/author, I could have my RAW.works edition wiki about that book, here on the site. Generative AI makes this easier than ever. So while that might have taken hours of time to do in the past, you could do it in a few minutes now. You could potentially even automate the creation of the wiki, pulling data from various internal and external locations to curate a "showcase" wiki entry for the book. What I'm excited about here is the possibility to create really engaging experiences that are more than just a recommendation and a link to go somewhere else. My next thought immediately terrified me. If you think filter bubbles are a problem now, this is gonna make them even worse. So for example, Google could functionally summarize the entire internet with AI in a way that you will never leave Google. And you can see that they're already doing this. With certain types of questions, you get the response as a snippet in the top search results of Google. (Not necessarily link to another website, but an official "Google answer.) You can imagine taking that one step further, where the user literally never leaves Google. So you start on Google and when you ask it for stuff, it doesn't give you the link to the other person's website. It gives you the summary that it wants you to see. I'm not sure Google will ever actually implement this to the fullest extreme, because all of their money comes from selling ads via the cost per click. The traffic leaving Google and going to other people's websites is their main cash cow business. But it's not hard to imagine a version of Google where there are no organic search results, only the Google AI summaries and paid ads. If not Google, someone with less ad revenue. Microsoft Bing I'm sure makes some money on ads, but it's not really the lifeblood of the organization. Or Amazon can create its own version of the internet that is constantly making sure that you see the version that maximizes as many of the physical and digital products that they sell as possible. Maybe we're already living in a version of Indra's web. In the context of books, Amazon is already the number one platform for physical books, digital books, audiobooks, and now they own Goodreads...so it's kind of hard not to be in Amazon's version of the web when you're talking about books. Interdependent. Interconnected. Addicted to shiny jewels. ### DALL·E 3 Can Almost Spell - URL: https://raw.works/dalle-3-can-almost-spell/ - Source: content/works/dalle3_can_almost_spell/index.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-04T00:00:00Z", "image": "/images/_654b07bb-cbc9-4723-8bae-2d65405c77f3.jpeg", "lastmod": "2023-10-04T00:00:00Z", "publishdate": "2023-10-04T00:00:00Z", "tags": [], "title": "DALL·E 3 Can Almost Spell" } ``` Content: As amazing as Midjourney is, the one thing it is notoriously bad at is spelling. It doesn’t understand words as text. When OpenAI recently [teased DALL·E 3 with their avocado therapy cartoon](https://openai.com/dall-e-3), my immediate reaction was “OMG it can spell!” Aside from some [goofy](https://www.linkedin.com/posts/raymondweitekamp_additivemanufacturing-3dprinting-additivemanufacturing-activity-7067498448454889472-ouk3?utm_source=share&utm_medium=member_desktop) [LinkedIn](https://www.linkedin.com/posts/raymondweitekamp_cyclic-olefin-resin-not-your-grandpas-photopolymer-activity-7067181095028461569-kmJj?utm_source=share&utm_medium=member_desktop) [posts](https://www.linkedin.com/posts/raymondweitekamp_3dprint-activity-7061539132195442688-wc7W?utm_source=share&utm_medium=member_desktop) - I was really struggling to see how Midjourney could be used for my business. It was a fun toy, and clearly a very powerful one, but it didn’t seem practical for business use. I’m sure that there are plenty of Midjourney diehards who are proving me wrong on a daily basis, but for me something about the lack of spelling just gave me the sense that Midjourney would be more trouble than it was worth for our professional digital marketing use cases. So the real question is: **can DALL·E 3 actually spell?** (As of this writing, the only way I know how to access it is through the [Bing Image Creator](https://www.bing.com/create), which makes me question my $20/mo OpenAI “ChatGPT Plus” subscription. (Microsoft must really have OpenAI by the balls.)) I’ll let you decide for yourself. (All typos courtesy Microsoft via Bing Image Creator) {{< gallery match="images/*" sortOrder="desc" rowHeight="300" margins="5" thumbnailResizeOptions="600x600 q90 Lanczos" showExif=false previewType="blur" embedPreview=true loadJQuery=true >}} ### Plumber.AI - URL: https://raw.works/plumber.ai/ - Source: content/works/plumber_ai.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-03T00:00:00Z", "lastmod": "2023-10-03T00:00:00Z", "publishdate": "2023-10-03T00:00:00Z", "tags": [], "title": "Plumber.AI" } ``` Content: Do you know whose job AI is not going to be replacing anytime soon? My plumbers. Literally every week there is a new advancement in generative AI that gets us one step closer to AI that can achieve the "Turing Test equivalent output" for many functional roles within an organization. A friendly agent to help you sift through the documentation and clarify any questions? You can [set that up in less than an hour](/my-first-executive-ai-win/). How about write a blog? That was so 2022. What about someone who could provide a quick sounding board to test your ideas, or an intern you can research different business opportunities? This is all easy to do with tools that are either free or cost as much for a monthly subscription as a human might charge per hour or even per 15 minutes. I have been following these developments relatively closely and I have yet to see an AI that can replace copper piping, install a water heater, or shut off the water if a pipe bursts under my house. While a huge number of programmers have been laid off here in the Bay Area, my plumber is doing just fine. ### Knowing About - URL: https://raw.works/knowing-about/ - Source: content/works/oct2.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-02T00:00:00Z", "lastmod": "2023-10-02T00:00:00Z", "publishdate": "2023-10-02T00:00:00Z", "tags": [], "title": "Knowing About" } ``` Content: With LLMs accelerating daily, the value of “knowing about” something is approaching zero. To use an example that is relevant to my business [polySpectra](https://polyspectra.com/?ref=rawworks): previously it might have been valuable to be the person on your team or in your company that knows about "resin 3D printing". In about 90 seconds I just tested (via [Poe](https://poe.com/?ref=rawworks)) GPT 4, Claude-2, Llama-2-70b, and PaLM 2. They all “know about” resin 3D printing (see below for examples). If someone knows how to spell the thing you are curious about - they can get a pretty good overview of a topic instantaneously and for free. So your value as the person who “knows about” the topic isn’t very high, because in the time it would take me to find you, I already have the answers I need. So how is a mere human supposed to stay ahead of the curve? Perhaps you can know something that is fundamentally unknowable to AI. Perhaps you can do something better and faster and cheaper than anyone else. (“Knowing how” is clearly infinitely more valuable than “knowing about”.) Or perhaps you can choose a few things or maybe even just one thing - you can choose a topic on which you are going to be the absolute foremost expert in the entire world. Not a topic that you “know about”, a topic that you know in your bones, a topic that you live every single day. Examples for fun: {{< blockquote author="GPT-4" >}} # Resin 3D Printing Resin 3D printing, also known as stereolithography (SLA), digital light processing (DLP), or liquid crystal display (LCD) printing, is a popular method of additive manufacturing that uses light to cure liquid resin into solid objects. ## How Does Resin 3D Printing Work? Resin 3D printing involves the use of a light source — UV laser in SLA, projector light in DLP, or LCD panel in LCD printing — to cure and solidify a photosensitive liquid resin layer by layer. The light source is controlled by a computer, which follows a digital 3D model to guide the light along specific paths for each layer. Here's a step-by-step breakdown of the process: 1. The build platform is lowered into the resin vat, leaving a thin layer of liquid resin between the build platform and the bottom of the vat. 2. The light source then traces the first layer of the 3D model, hardening the resin where it strikes. 3. The build platform moves up to peel off the hardened layer from the bottom of the vat and allows new resin to flow beneath. 4. The process repeats, layer by layer, until the object is complete. Each new layer adheres to the one beneath it, resulting in a complete 3D object. 5. After printing, the object is typically cleaned in a solvent to remove excess resin and then post-cured under UV light to achieve full hardness and mechanical properties. ## Materials The primary material used in resin 3D printing is a type of photopolymer known as resin. The specifics of the resin, such as its color, hardness, and finish, can vary significantly. Some common types include: - Standard resins: Good for general purposes and easy to print with. - Tough and durable resins: Designed to withstand higher stress and strain, mimicking the properties of materials like ABS or polypropylene. - Flexible resins: Can mimic the properties of materials like rubber. - Castable resins: Burn out cleanly from molds, ideal for jewelry or parts that will be cast in metal. - High-detail resins: Designed to capture very fine details, often used in modeling or dentistry. ## Advantages of Resin 3D Printing - **High resolution and accuracy:** Resin 3D printers can produce parts with a high level of detail and precision, making them ideal for applications that require intricate designs, such as jewelry, dentistry, and model making. - **Smooth surface finish:** Objects printed using resin 3D printers typically have a very smooth surface finish compared to other forms of 3D printing, reducing the need for post-processing. - **Material variety:** There are many different types of resins available, each with unique properties, allowing for a wide range of applications. ## Disadvantages of Resin 3D Printing - **Post-processing:** While the surface finish is typically smooth, all parts require cleaning and post-curing after printing, which can be messy and time-consuming. - **Material handling:** Resins can be sticky, messy, and often require careful handling due to their sensitivity to UV light. They can also be potentially harmful or allergenic, requiring protective equipment when handling. - **Size limitations:** Resin 3D printers typically have smaller build volumes than FDM printers, making them less suitable for large parts. - **Cost:** The cost of resin can be high, and the printers themselves can also be more expensive than FDM printers. ## Applications of Resin 3D Printing Resin 3D printing is used in a variety of industries and applications, including: - **Dental and medical:** Resin printers can produce highly accurate dental models, surgical guides, and prosthetics. - **Jewelry:** The high precision and smooth surface finish make resin 3D printing popular in jewelry making, especially combined with castable resins. - **Prototyping and product design:** Resin 3D printing is great for producing detailed prototypes that closely resemble the final product. - **Model making and miniatures:** The high-resolution capabilities make resin 3D printing a popular choice for creating highly detailed models and miniatures, such as those used in tabletop gaming. {{< /blockquote >}} {{< blockquote author="Claude 2" >}} # Resin 3D Printing - Resin 3D printing, also known as stereolithography (SLA) or vat photopolymerization, uses liquid photopolymer resins that are cured and solidified by light to build objects layer-by-layer. - A UV laser or projector selectively cures and solidifies each layer of resin in the build area based on the 3D model. The build platform then raises out of the vat of resin to allow fresh liquid resin to flow under and prepare for the next layer. - Common resin materials include epoxy and acrylate-based photopolymers. The resins are designed to cure rapidly with UV light exposure. - Resin 3D printers offer very high resolution and detail, with layer thicknesses as low as 25-100 microns. Smooth surfaces and fine features can be produced. - Supports are needed during printing to prevent deformation by holding up overhangs and spanning gaps in the model. Supports are removed after printing is complete. - Resin printing is good for smaller, high detail prints like jewelry, miniatures and dental models. The build area is typically smaller than other 3D printing methods. - Resins can have limitations in terms of strength, heat resistance and UV stability compared to other materials. Post-curing under UV is often used to improve mechanical properties. - Resin printers require cleaning and post-processing to remove excess liquid resin after printing. Isopropyl alcohol is commonly used for this. - Resin tanks need periodic replacement as the transparent bottom wears out over time. Resin also has a shelf life and expires after 6-12 months typically. {{< /blockquote >}} {{< blockquote author="Llama-2-70b" >}} # Resin 3D Printing Resin 3D printing is a type of 3D printing technology that uses a photopolymer resin as the printing material. The resin is a light-sensitive material that hardens when exposed to light, allowing for precise and detailed prints. ## Printing process The printing process involves a layer of resin being exposed to light through a mask or laser, which hardens the resin in a specific pattern. The build platform then moves down, and the process is repeated, layer by layer, until the entire print is completed. ## Photopolymer resin The photopolymer resin is a light-sensitive material that hardens when exposed to light. It is typically made up of a mixture of a resin base and a photoinitiator, which is responsible for the hardening process. ## Resolution Resin 3D printing is capable of producing prints with high resolution and detail, typically in the range of 10-50 microns. This makes it ideal for printing small, intricate parts and models. ## Accuracy Resin 3D printing is known for its accuracy and precision, with prints often being within 0.1-0.2mm of the original design. ## Layer thickness The layer thickness in resin 3D printing can vary depending on the printer and the specific resin being used. Typical layer thicknesses range from 10-100 microns. ## Post-curing After the print is complete, it may require post-curing, which involves exposing the print to light for a longer period of time to fully harden the resin. ## Finishing Resin prints can be finished in a variety of ways, including sanding, priming, and painting. ## Cost Resin 3D printing is generally more expensive than other 3D printing technologies, due to the cost of the photopolymer resin and the specialized equipment required. ## Build volume Resin 3D printers typically have smaller build volumes compared to other 3D printing technologies, due to the need for precise light exposure. ## Applications Resin 3D printing is commonly used for printing small, intricate parts and models, such as in jewelry, dentistry, and engineering. It is also used for prototyping and artistic applications. Some popular resin 3D printers include: * Formlabs Form 3 * Prusa i3 * Photon S Some popular resin 3D printing software includes: * Formlabs PreForm * PrusaSlicer * Photon Slicer * Cura * Slic3r Some popular resin 3D printing materials include: * Formlabs Standard Resin * Formlabs Grey Resin * Formlabs Clear Resin * Prusa Resin * Photon Resin * MakerGeeks Resin {{< /blockquote >}} {{< blockquote author="PaLM 2" >}} ## Resin 3D Printing Resin 3D printing, also known as stereolithography (SLA), is a 3D printing technology that uses a laser to cure liquid resin into solid objects. The process starts with a 3D model, which is sliced into thin layers. The laser then traces each layer of the model onto the surface of a vat of resin, curing the resin and solidifying it into a solid object. There are two main types of resin 3D printers: * **DLP (Digital Light Processing)** printers use a projector to cure the resin. DLP printers are faster than SLA printers, but they produce less detailed objects. * **SLA (Stereolithography)** printers use a laser to cure the resin. SLA printers are slower than DLP printers, but they produce more detailed objects. Resin 3D printing is used for a variety of applications, including: * **Prototyping**: Resin 3D printing is a fast and cost-effective way to create prototypes of new products. * **Manufacturing**: Resin 3D printing can be used to manufacture small batches of custom parts. * **Art**: Resin 3D printing can be used to create unique and intricate works of art. Here are some of the advantages of resin 3D printing: * **High accuracy**: Resin 3D printers can produce objects with very high accuracy. * **Smooth surface finish**: Resin 3D printed objects have a smooth surface finish, which makes them ideal for applications where aesthetics are important. * **Wide range of materials**: There is a wide range of resins available for resin 3D printing, which allows users to create objects with different properties. Here are some of the disadvantages of resin 3D printing: * **Slow speed**: Resin 3D printers are slower than other 3D printing technologies, such as FDM (Fused Deposition Modeling). * **High cost**: Resin 3D printers and resins can be expensive. * **Post-processing**: Resin 3D printed objects often require post-processing, such as cleaning and curing. Overall, resin 3D printing is a versatile and powerful technology that can be used for a variety of applications. However, it is important to be aware of the advantages and disadvantages of resin 3D printing before deciding if it is the right technology for your needs. {{< /blockquote >}} ### Professional 'Advisors' - URL: https://raw.works/professional-advisors/ - Source: content/works/professionaladvisors.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-10-01T00:00:00Z", "lastmod": "2023-10-01T00:00:00Z", "publishdate": "2023-10-01T00:00:00Z", "tags": [], "title": " Professional 'Advisors' " } ``` Content: There is something that always bothered me about the advice of productivity gurus. It has taken me a while to figure out what it is. I think a big part of it is that I was never particularly impressed with their examples. A common example, especially when someone is writing a productivity book, would be examples of authors. The first problem with this is that I'm not particularly interested in writing a book (although maybe someday I will be). More over, the specific act of writing a book is not necessarily difficult or impressive. It is definitely a lot of work to write a book. (Or was, before generative AI took off.) And I'm sure that if you actually want anyone to read your book, it is even more work. And if you actually want to make the book successful enough to make any money, it is even more work. But the example of the author who just writes for some specified period of time or some specified number of pages every single day really didn't resonate with me. Maybe the other piece of it is that the majority of the people giving advice on productivity themselves aren't doing anything or achieving anything that I find particularly impressive or applicable to my own life. Especially where their profession now is simply writing books about methods for how to be more efficient at writing books. Again, I'm sure that is valuable for a lot of aspiring authors, but it never really stuck with me. For me, I have always been more interested in the advice of operators and doers. For example, I am much more curious about Warren Buffett's advice on investing because he is actively investing and has built up his expertise on the topic through doing it (well) for decades. His advice might get him a little bit of publicity, but it really isn't his business model to be giving advice. His business model is to be a smart investor. On the other hand, you have people like Jim Cramer, whose primary job is giving investment advice every single day on TV. That's what he gets paid to do. And yes, he also had a hedge fund at one point. I won't pretend to know anything about it, but I don't think it was particularly successful. Regardless, his profession is giving advice about investing, not investing. I tend to be more interested in productivity tips from people who are focused on doing and achieving things, not just talking about it. The hard-won learnings of an operator are richer and more rooted in reality. {{< blockquote author="Ram Dass" link="https://amzn.to/3PGuZQq" title="Be Here Now" >}} *I can do nothing for you but work on myself...you can do nothing for me but work on yourself!* {{< /blockquote >}} ### Why Does Handwriting OCR Suck in 2023? - URL: https://raw.works/why-does-handwriting-ocr-suck-in-2023/ - Source: content/works/handwritingOCR.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-30T00:00:00Z", "lastmod": "2023-09-30T00:00:00Z", "publishdate": "2023-09-30T00:00:00Z", "tags": [], "title": "Why Does Handwriting OCR Suck in 2023?" } ``` Content: Some days, I'm overwhelmed by how fast technology is improving. Today, I feel like I'm running Windows 98. [In 1998 "50% of all handwritten card and letter addresses were 'read' not by human eyes but by high speed sorting machines, a process that resulted in labor savings of $31 million."](https://www.buffalo.edu/cubs/research/pioneers-in-ai-systems.html) The fundamental challenge of a machine reading handwritten postal addresses was solved before 1998. So why can't even the most advanced AI algorthims from the top tech companies read a handwritten note in 2023? (This is not a rhetorical question.) Optical character recognition (OCR) for digital text is ubiquitous, free, and fast. The built-in camera app on your iPhone will scan digitally-printed text basically in real-time. Google Translate will overlay the translation of signs/receipts/packaging in augmented reality. Ironically, one of the [classic tutorials](https://www.digitalocean.com/community/tutorials/how-to-build-a-neural-network-to-recognize-handwritten-digits-with-tensorflow) in machine learning education is recognizing handwritten digits. Apparently only the US Postal Service has the talent to put this into production. Either that, or no one can think of a way that handwriting OCR would make them money. An efficient way to digitize pen and paper means that people would spend less time on their phones/computers/tablets, which would mean fewer notifications and less ad revenue. It would also mean that a plain piece of paper might displace the need for a $1500 tablet with a cool "pencil" to boot...and that's bad for hardware sales. (Not to mention less data to spy on.) I've been obsessed with finding the best tools for handwriting OCR, as a way of digitizing my analog notes (with the goal of reducing my screen time). Ironically, my journey thus far has only added to my screentime, trying to find and try "handwriting to text" tools that might unlock this workflow for me. **My conclusion as of today:** the best handwriting OCR tools are already built into your phone. For iOS, use the "Scan Documents" feature inside of Notes. For Android, it's called Google Lens. Don't waste your time or money on any of the other apps for this --- they all suck. Adobe gets a dishonorable mention for how terrible their OCR is for handwriting. (Given that they have monopolized PDF and "regular" OCR forever, they should be the best, not the worst.) If you find a handwriting OCR app that doesn't suck, please contact me immediately. I am hoping that maybe in 15 more years, I can finally digitize my handwritten notes. ### Stories <> Data - URL: https://raw.works/stories-data/ - Source: content/works/stories_data.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-29T00:00:00Z", "lastmod": "2023-09-29T00:00:00Z", "publishdate": "2023-09-29T00:00:00Z", "tags": [], "title": "Stories \u003c\u003e Data" } ``` Content: Humans want stories, not data. Algorithms want data, not stories. The major breakthrough of LLMs is commuting between these two domains. ### Don't Get X'd - URL: https://raw.works/dont-get-xd/ - Source: content/works/dontgetxd.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-28T00:00:00Z", "lastmod": "2023-09-28T00:00:00Z", "publishdate": "2023-09-28T00:00:00Z", "tags": [], "title": "Don't Get X'd" } ``` Content: I am proposing yet another definition of X. It is a verb this time. You know that you've been X'd when the platform that you've been [sharecropping](/sharecropping) on suddenly pulls the rug out from under you. Are you familiar with [Nassim Taleb](https://fooledbyrandomness.com/?ref=rawworks)'s parable of the turkey? If not, here's a quick summary from my friend Claude: {{< blockquote >}} *The parable goes that a turkey is fed by a farmer every day for 1000 days. Each day confirms the turkey's belief that the farmer cares for it and will continue feeding it. On day 1001, right before Thanksgiving, the farmer kills and eats the turkey.* *The point is that the turkey suffered from confirmation bias and misunderstood its true environment. Just because something has worked numerous times in the past doesn't mean it will necessarily continue working that way in the future. The turkey made the mistake of inductively inferring a rule from a set of observations that were not representative of the full set of possible observations.* *So in essence, the parable is a cautionary tale about the problem of induction, optimism bias and believing the future will resemble the past without enough evidence to support that belief. It's meant to encourage wariness of black swan events - those that lie outside the realm of regular expectations but carry extreme impact.* {{< /blockquote >}} (If you haven't read it, I highly recommend [The Black Swan](https://amzn.to/3LFqTXK).) There's also the story about the [highly profitable McDonald's at the top of the volcano](https://tim.blog/2023/07/19/bill-gurley-interviews-tim-ferriss/?ref=rawworks). (It's a very short story.) I'm not recommending that everyone switch to running Arch Linux. As they say, *"Linux is only free if your time is worth nothing"*. In fact, these days [I'm not really trying to give recommendations](/advice-under-uncertainty/). Instead of advice, here are some questions I'm asking myself, in an attempt to not get X'd. More accurately, to make f(x) less painful when I inevitably do get X'd. ([RWRI](https://realworldrisk.com/?ref=rawworks) joke) - Do I own it? (Am I paying for it? What exactly do I own?) - Can I export all of my data? (And how painful is that operation? And have I actually tried practicing getting my data out? ) - What is the "true" business model of the company making the product/platform? - How much will my life suck if Elon Musk buys the company that makes this product? - How long has it been around? ([Lindy's Law](https://en.wikipedia.org/wiki/Lindy_effect)) ### Introducing Rai, my AI Librarian - URL: https://raw.works/introducing-rai-my-ai-librarian/ - Source: content/works/introducingRai.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-27T00:00:00Z", "lastmod": "2023-09-27T00:00:00Z", "publishdate": "2023-09-27T00:00:00Z", "tags": [], "title": "Introducing Rai, my AI Librarian" } ``` Content: As you might already know, I'm having a lot of fun with AI right now. It took me [4 hours to set up my customer support chatbot](/my-first-executive-ai-win/) for [polySpectra](https://polyspectra.com). It took me 40 minutes to set up Rai, my new AI librarian for RAW.works. (Next time I'm aiming for 4 minutes.) Put Rai to the test! That's the only way it (they?) will improve. P.S. - I want to [be like Seth Godin](https://seths.blog/2023/03/opening-the-pod-bay-door/) when I grow up. ### What If The Web Worked For You? - URL: https://raw.works/what-if-the-web-worked-for-you/ - Source: content/works/whatifthewebworkedforyou.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-26T00:00:00Z", "lastmod": "2023-09-26T00:00:00Z", "publishdate": "2023-09-26T00:00:00Z", "tags": [], "title": "What If The Web Worked For You?" } ``` Content: The largest tech companies in the world are selling your attention. At the same time, the internet is the great democratizer, and is giving you access to the vast majority of the world at lightning speed. You can't possibly read everything on your many screens, and you can't possibly disconnect completely. What's a human to do? This is a very big topic. Today, I'm particularly curious about "browser automation" as a way to create/enforce my own "terms of engagements" with the internet. Silencing notifications is great defense. [Automating with APIs or schedulers is cool](/automating-social-media-posts-for-hugo-sites/). But the reality is that many of the most valuable corners of the internet are out of reach unless you are logged into the app. Why? Business. (See Sentence 1.) Why did Google kill Reader? It gives you too much control of your attention, and they need to sell your attention. After completely forgetting about RSS feeds for over a decade (after Google Reader was killed), I am currently testing [Feed.ly](https://feedly.com/) as a potential mode of "browser automation". With a subscription, you can even use it to sign up for email lists, to keep newsletters out of your inbox. I recently got curious about [Buffer](https://buffer.com/) as a way to [automatically post to social media](/automating-social-media-posts-for-hugo-sites/) (without me needing to log in.) Even though most social media platforms have an API and from a technical perspective this should be really easy - the social media companies simply don't want to make it easy to automate posts. This makes sense, because then their job keeping spam out gets harder. But I simply don't want to get sucked into a LinkedIn feed rabbit hole every time I want to make a post...and as easy as it sounds to *"just have self-control"* - I can't tell you how many times I've logged into to post one thing and then woken up an hour later hating myself and seriously doubting my self-worth. (My feed is full of the amazing accomplishments of my friends and peers, which is certainly nice in modest doses, but not something I can easily stop scrolling.) Not only are APIs annoying to sort out (and even with Buffer it's not really just "set it and forget it") - but they change at the whims of their owners. Around the same time that Twitter changed their name to X, they also deprecated a bunch of really useful API features. The changes were so dramatic that [some companies needed to shut down completely](https://leaderbird.co/#note). Which brings me to "browser automation". I'm a complete n00b. I haven't set up anything cool yet, but I am very curious about trying to use some of these tools as a way of protecting my attention while engaging with parts of the internet that require a "human" UX. The two best tools I've found so far are: [Axiom.AI](https://axiom.ai/a?afmc=5j) & [Browse.AI](https://browse.ai/?via=raymond-weitekamp). (Both are my affiliate links which give you a small discount.) Both [Axiom.AI](https://axiom.ai/a?afmc=5j) & [Browse.AI](https://browse.ai/?via=raymond-weitekamp) promise to be able to "create an API for any website". Right now I'm not particularly interested in scraping, but both seem great for that use case. As a first use case, I'm curious to see if I can just follow [this example](https://axiom.ai/recipes/automate-facebook-posts) to post to Facebook from a Google Sheet. (Having an app log into Facebook on my behalf to make the post.) More to come as I start experimenting! In the meantime, what strategies have you found helpful for gaining more control over your online experiences? ### CO2 Offsets for Walking? Why Not? - URL: https://raw.works/co2-offsets-for-walking-why-not/ - Source: content/works/treecard.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-25T00:00:00Z", "lastmod": "2023-09-25T00:00:00Z", "publishdate": "2023-09-25T00:00:00Z", "tags": [ "1st Time" ], "title": "CO2 Offsets for Walking? Why Not?" } ``` Content: Yesterday, I found out about [TreeCard](https://treecard.onelink.me/0db8/5pwci33h), an app from [Ecosia](https://www.ecosia.org/) that pays to plant trees based on how many steps you take. Today I'm at a 95-days streak of walking 10,000 steps a day - so my first reaction was that I wished I had found out about this a few months ago! (Why so many steps? I read [Built to Move](https://www.amazon.com/Built-Move-Essential-Habits-Freely-ebook/dp/B0B5STDSC8?crid=2RVRE8P5FDY07&keywords=built+to+move&qid=1695675631&sprefix=built+to+move%2Caps%2C1835&sr=8-1&linkCode=ll1&tag=rawwerks09-20&linkId=c6f6247231493eef6b2eac6ba98cf73e&language=en_US&ref_=as_li_ss_tl).) My second reaction: Where does the money come from? Here's the official TreeCard answer: {{< blockquote >}} *Brands pay us to feature their eco products within our rewards structure. These can be earned by planting trees!* {{< /blockquote >}} So basically they're selling my attention. (What's new?) I'm not sure that this alone is going to save the planet, but so far my review is: _**why not?**_ I'm going to get those steps anyways, why not get some carbon offsets along the way? I'm still figuring out how the app works, but apparently if you join with [my Treecard link](https://treecard.onelink.me/0db8/5pwci33h) and enter the code `ray-5d7` then we both get extra trees (or something like that). ### A Recipe for Forgiveness - URL: https://raw.works/a-recipe-for-forgiveness/ - Source: content/works/arecipeforforgiveness.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-24T00:00:00Z", "lastmod": "2023-09-24T00:00:00Z", "publishdate": "2023-09-24T00:00:00Z", "tags": [], "title": "A Recipe for Forgiveness" } ``` Content: - **One holy day** - Optionally in your tradition of faith, but someone else’s holy day will do - **[One patio chair](https://amzn.to/46iNOzU)** - Set up your patio chair in a peaceful location - **[One decorative pillow](https://amzn.to/3RxcfFC)** - Place the decorative pillow on the chair for added comfort - **[15mg Ketamine HCl](/a-joyous-equinox/)** - **[iPhone](https://amzn.to/3LBlmkK) with [AirPods](https://amzn.to/3LAHnQT)** - **Playlist** including but not limited to: - [**"Sit Around the Fire "** by Jon Hopkins with Ram Dass, East Forest](https://open.spotify.com/track/5kdXiiF8MxyLVhRjkVv9jQ) - [**"Indian Summer"** by Jonsi & Alex Somers](https://open.spotify.com/track/5IjIY6VMulVaAhBnMnDXbl?si=ab59735d06694a24) - **At least one felt sense of unworthiness**, preferably from the present - Finely dice the unworthiness into the smallest possible pieces - **At least one memory of being scolded for “bad” things you did as a child** - Whisk this memory thoroughly and compassionately - (Optionally) **tears** - Cry to taste - **A shower** - As long as necessary ### A Joyous Equinox - URL: https://raw.works/a-joyous-equinox/ - Source: content/works/ajoyousequinox.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-23T00:00:00Z", "featuredimage": "/images/joyous-solstice.jpg", "image": "/images/joyous-solstice.jpg", "lastmod": "2023-09-23T00:00:00Z", "publishdate": "2023-09-23T00:00:00Z", "tags": [ "1st Time" ], "title": "A Joyous Equinox" } ``` Content: On the day of the equinox, I found myself embarking on a new journey, one that I am now eager to share. A great occasion for a pinch of discovery and a cup of self-care. This was my Day One trying out [Joyous](https://www.joyous.team/?ref=rawworks), a brand of low-dose ketamine therapy that you can self-administer within the comfort of your own home. The Day One recommended dose is 15mg, to make sure that the medicine is well tolerated. I chose the peppermint flavor of troche, which I found non-offensive and maybe even slightly fun. The text from [Joyous](https://www.joyous.team/?ref=rawworks) suggested that I set aside 20 minutes and set an intention. A few of the suggestions: *"Do you want to feel calm? Feel self-love? Experience insights? Forgive?".* I chose "feel self-love". (I thought maybe I'd save "forgive" for Yom Kippur tomorrow.) With the equinox sun casting long shadows in my backyard, I put a chair on the lawn to create a comfortable spot to sit and meditate. Right after putting the tiny white cube between my gum and cheek, I set a timer for 17 minutes, a duration that usually marks the boundary of my meditation comfort zone, the point at which I usually start to squirm. In addition to the "self-love" intention, I set the intention to just remain in seated meditation until the timer chimed, open to the experience and embracing whatever feelings and insights might arise. As I closed my eyes and focused on my breath, I felt an immense sense of calm wash over me. A feeling of "everything in its right place". It was a profound reassurance, a gentle whisper in the back of my mind that said, "This is all I need." In this moment, I felt connected to the world around me in a way that was deeply grounding. It seemed a perfect way to celebrate the equinox, a time of balance and harmony. The world outside and within me felt synchronized. I was completely present, not anxiously racing towards the future or entangled in the web of old memories. I was here, now, and that was all that mattered. As usual, I got pretty squirmy around 15 minutes, but stayed seated until the timer went off. In this slightly dissociated state, something about the timer made me think I needed to wrap everything up. I brought the chair back up to the deck, but as I stood up, I felt a bit dizzy. Upon reflection, I realized that I should have allowed myself to stay seated for a while longer, to let the experience fully run its course and to gently transition back into my usual state of consciousness. In retrospect, the synchronicity of the [Joyous](https://www.joyous.team/?ref=rawworks) package arriving on the day of the equinox feels like more than just a coincidence. The equinox, a time when day and night hold equal sway, seems like the perfect moment to commence a journey towards inner balance and harmony. I am grateful for this fun alignment, for the opportunity to begin this exploration on such a significant day. Re: [Joyous](https://www.joyous.team/?ref=rawworks) - my curiosity is piqued. I am eager to continue exploring the therapeutic potential of low-dose ketamine, as a tool to enhance my overall wellbeing. ### Balancing Quantity and Quality in Content Creation - URL: https://raw.works/balancing-quantity-and-quality-in-content-creation/ - Source: content/works/qualityandquantity.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-22T00:00:00Z", "lastmod": "2023-09-22T00:00:00Z", "publishdate": "2023-09-22T00:00:00Z", "tags": [ "" ], "title": "Balancing Quantity and Quality in Content Creation" } ``` Content: In the chaotic landscape of the digital age, content is king. Companies, individuals, and brands compete in a relentless bid for attention, and content creation has become the currency of this competition. The sheer volume of information available at our fingertips is staggering. We are bombarded with blogs, podcasts, videos, social media posts, and more, all vying for our time and attention. In this whirlwind of content creation, a dilemma often surfaces: should the focus be on [quantity or quality](/quantity-over-quality)? On one hand, producing a large amount of content regularly can improve visibility, boost Search Engine Optimization (SEO), and keep an audience engaged. On the other hand, high-quality content can build credibility, drive conversions, and create loyal followers. But, can one really strike a balance between the two or does one inevitably sacrifice the other? In this post, we will delve into the tug of war between quantity and quality in content creation, examining the importance of each, the struggle to balance the two, and strategies to maintain this balance effectively. ## The Importance of Quantity In the digital age, quantity plays a crucial role in content creation. Producing content frequently and consistently can help establish an active online presence and foster an ongoing relationship with your audience. Quantity can also serve as a catalyst for improving visibility. Each piece of content you create is a new opportunity for search engine indexing, which in turn can bolster your Search Engine Optimization (SEO). More content equals more keywords, leading to better visibility in search results. Furthermore, producing content regularly keeps your audience engaged. It creates an expectation among your followers, who come to understand that they can consistently find fresh, interesting material on your site or social media platforms. Consider the success of BuzzFeed. This media company has effectively harnessed the power of quantity, publishing an extensive range of content daily, from news articles to quizzes, videos, and listicles. This strategy has not only attracted a vast audience but also helped maintain high engagement levels. However, it's crucial to note that a "quantity-only" approach may lead to diminishing returns. With the internet flooded with content, consumers have become more discerning. They are not just seeking more content; they are seeking content of value. An overemphasis on quantity can lead to a dilution of quality, which may result in content that feels rushed, repetitive, or lacking in depth. Quantity, without the balance of quality, risks turning into mere noise in an already crowded digital space. As such, while quantity plays a significant role in content creation, it should not come at the expense of quality. The two must work in tandem to create truly compelling and effective content. ## The Significance of Quality While quantity can improve visibility and keep your brand at the forefront of your audience's mind, quality is what truly defines your brand's voice and image. Quality content is not just about producing well-written, error-free articles or aesthetically-pleasing visuals. It's about delivering value to your audience, addressing their needs, and providing insights that they can't find elsewhere. Quality content can build credibility and establish your brand as a thought leader in your industry. It can help you stand out in an ocean of mediocre content. When you consistently produce high-quality content, you demonstrate to your audience that you value their time and aim to provide them with useful, engaging information. Moreover, quality content can drive conversions. When your audience trusts your brand and finds value in your content, they are more likely to convert into customers. High-quality content can also foster loyalty, encouraging customers to come back for more and even become brand ambassadors. For instance, The New Yorker has long been revered for its high-quality content. Each published piece, whether it's a long-form article or a short commentary, is meticulously crafted, providing in-depth analysis and unique perspectives. This commitment to quality has not only built an incredibly strong brand reputation but also a loyal readership that values and trusts its content. Apple, as a brand, is another prime example. Their product announcements and marketing campaigns are always highly anticipated due to their reputation for delivering high-quality, innovative content that offers more than just product specifications. They tell a story, evoke emotions, and create a unique experience for their audience. This focus on quality has helped them build a loyal customer base and a powerful brand. However, while the importance of quality is clear, consistently producing top-quality content can be a significant investment. It requires a considerable amount of time, effort, and resources. High-quality content often involves extensive research, careful planning, intricate design, multiple revisions, and a deep understanding of your audience's needs and preferences. This level of detail and dedication in content creation can quickly become expensive, especially for smaller businesses or individuals operating on a limited budget. This is where the balance between quality and quantity comes into play. While striving for the best quality, it's important to also consider the practical implications, including the cost and time investment, without compromising so much that the content loses its value and purpose. ## The Struggle Between Quantity and Quality Finding the right balance between quantity and quality is a common challenge that many content creators grapple with. It's like walking a tightrope, where leaning too much towards either side can cause a fall. On one side, there's the pressure to produce a large volume of content to stay visible and relevant. With the fast-paced nature of the digital world, content can quickly become outdated, and there's always a demand for fresh material. The fear of being forgotten can drive creators to churn out content constantly, sometimes at the expense of quality. On the other side, there's the desire to craft high-quality content that provides value, builds credibility, and fosters loyalty among the audience. This often involves thorough research, thoughtful analysis, careful editing, and a deep understanding of the audience's needs and preferences. However, this level of quality can't be achieved in haste and requires a significant time investment. As a content creator myself, I've faced this dilemma numerous times. I've found myself staying up late to fine-tune a blog post or spending hours editing a video to ensure it meets my quality standards. Conversely, I've also experienced the pressure of meeting content deadlines and the anxiety that comes with seeing my content calendar empty. The struggle between quantity and quality is real, and it's a challenge that requires careful consideration and strategic planning. But with the right approach, it's possible to strike a balance that maximizes the benefits of both. ## Practical Strategies for Balancing Quantity and Quality Walking the tightrope between quantity and quality is a nuanced dance that I have come to understand well. As I emphasized in my blog post "[Quantity Over Quality](/quantity-over-quality)", navigating this complex terrain requires strategic thinking and practical approaches. Here, I share some of the strategies I've found effective. 1. Embrace Iterations Perfectionism can be a roadblock on the path to consistent content creation. Instead, I've found it beneficial to focus on creating a 'good enough' version and refining it over time. This iterative approach ensures a steady flow of content, which can then be improved based on audience feedback and fresh insights. 2. Prioritize Your Content In my experience, not every piece of content demands the same level of attention and time. It's crucial to identify which content provides the most value to your audience and dedicate your resources accordingly. For content that's less critical, look for ways to create it more efficiently while still maintaining its core value. 3. The Power of Automation As I've detailed in my post "[Automating Social Media Posts for Hugo Sites](/automating-social-media-posts-for-hugo-sites/)", automation can be a content creator's best friend. Tools like Hootsuite, Buffer, and IFTTT can take the burden of scheduling and managing social media posts off your shoulders, allowing you to concentrate on producing high-quality content. 4. Repurpose Your Content Creating versatile content that can be transformed into different formats is a strategy I've found particularly beneficial. A comprehensive blog post, for instance, can be repurposed into a podcast episode, a video, or a series of social media posts. This approach allows you to maximize your content's reach and visibility without the need for continual new content creation. 5. Leverage AI and Machine Learning Tools I've seen firsthand the power of AI and machine learning in enhancing content creation efficiency. As I shared in my post "[My First Executive AI Win](/my-first-executive-ai-win/)", these tools, like Grammarly for writing and editing or chatbots for customer support, can significantly reduce the time spent on non-strategic tasks. In the end, the act of balancing quantity and quality is an ongoing process, and the strategy that works best may vary based on your audience, resources, and objectives. The key is to remain adaptable, learn from your experiences continually, and always keep the needs of your audience at the heart of your content strategy. ## Conclusion Navigating the delicate interplay between quantity and quality in content creation is a challenge that every creator faces. As I've emphasized throughout my body of work, both elements have their unique and indispensable value. Quantity ensures consistency, fosters a regular connection with your audience, and allows you to iterate and improve. It's about showing up, delivering, and learning from the act of creation itself. The more you create, the more opportunities you have to hone your craft and engage with your audience. Quality, on the other hand, is what sets your content apart. It's the depth, the insight, and the value you provide that resonates with your audience and keeps them coming back for more. High-quality content reflects your brand's credibility and commitment to delivering value. However, the secret to successful content creation doesn't lie strictly in choosing one over the other. Instead, it's about finding your own balance and rhythm. It's about understanding your capacity, your audience's needs, and the resources at your disposal. It's about leveraging tools and strategies that can help you maintain a steady flow of quality content. In the end, successful content creation is a dance between quantity and quality, and every creator must find their own rhythm. As you continue on your content creation journey, I encourage you to embrace this dance. Experiment, learn, iterate and most importantly, enjoy the process. Remember, the goal is not perfection, but continuous growth and connection with your audience. ### Sharecropping - URL: https://raw.works/sharecropping/ - Source: content/works/sharecropping.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-21T00:00:00Z", "image": "/images/itunes-useragreement-southpark.jpg", "lastmod": "2023-09-21T00:00:00Z", "publishdate": "2023-09-21T00:00:00Z", "tags": [], "title": "Sharecropping" } ``` Content: Who gets paid for your work? Elon Musk? Alphabet shareholders? Medium? Meta? Does it bother you that you are a sharecropper? {{< figure src="/images/itunes-useragreement-southpark.jpg" title="You didn't read the iTunes User Agreement?" >}} ### My First Executive AI Win - URL: https://raw.works/my-first-executive-ai-win/ - Source: content/works/myfirstexecutiveAIwin.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-20T00:00:00Z", "lastmod": "2023-09-20T00:00:00Z", "publishdate": "2023-09-20T00:00:00Z", "tags": [ "buffer", "_buffer" ], "title": "My First Executive AI Win" } ``` Content: Most of my experience with machine learning and artificial intelligence has been in art. (More on that below, for the curious.) Today I'm excited to share my first "win" using AI as a business executive: [Mendable](https://www.mendable.ai/?ref=rawworks). I can't really take much credit other than to say that I've been keeping my eyes open, but the impact that Mendable will have on [polySpectra](https://polyspectra.com/?ref=rawworks) is going to be truly transformative. **TL;DR** - with [Mendable](https://www.mendable.ai/?ref=rawworks) I was able to train, test, and deploy a customer support chatbot on our full technical and product documentation in about 4 hours (and I will explain below how it could have been 40 mins.) If you run a business, have someone on your team try it today...you'll thank me. To put it simply, my team and I spend way too much time answering customer questions that are clearly in the documentation. *"What printers can I use with COR?", "What is the UTS of COR Black?", etc.* It's not a great use of time for our team, but it's clearly a necessary part of bringing prospects and customers up to speed. I'm as guilty as anyone for not reading the instructions before opening something. (Including some close calls where I really should have read the MSDS before opening the package.) I'm not blaming our customers - we're all under duress with today's information overload. Not reading is simply necessary for survival. But in manufacturing, optical equipment, and chemical safety - it is very important to read the documentation. So I had been on the hunt for a way to leverage the new developments in LLMs to help offload some of this burden. I also hate Intercom and other "dumb" chatbots - so I didn't want to implement something that I wouldn't be excited to use myself. A true win-win would be a chatbot that is helpful and properly trained on the product details and documentation, to help our customers get the information that they need in a matter of seconds, while keeping those "trivial" customer questions out of our inboxes. (The humans will answer the tough questions, for now.) A major hat-tip to [Mayo Oshin](https://www.siennaianalytics.com/?ref=rawworks) for turning me onto [Mendable](https://www.mendable.ai/?ref=rawworks) (via his [execellent newsletter](https://mayo-a.ck.page/59b4fbfc99)). I was actually considering learning LangChain myself to try to build out this idea. In this case, I'm incredibly grateful that someone built it first. Starting the clock at when I clicked the link to [mendable.ai](https://www.mendable.ai/?ref=rawworks) from Mayo's email, the sign up took a few minutes and I had successfully trained it on the product information from our Shopify account in less than an hour. Because we use Cloudflare to host our website, I ran into a technical hiccup where the Mendable bot was being blocked from scraping our documentation site from the sitemap, which thankfully the Mendable team was quickly able to work through with me over a couple of hours. Between lunch and dinner, I went from cold traffic to fully-onboarded customer running the app live on [docs.polySpectra.com](https://docs.polyspectra.com/?ref=rawworks). I was really impressed with the results from my test interactions with the bot, without any fine-tuning at all. I would encourage you to [put my bot to the test](https://docs.polyspectra.com/?ref=rawworks)! In summary, I am currently the #1 fanboy of [Mendable](https://www.mendable.ai/?ref=rawworks). --- For the curious (& because I promised above): My first machine learning application was for a composition for the Princeton Laptop Orchestra (PLOrk) entitled [G](http://plork.deptcpanel.princeton.edu/PLOrk-Spring2010/), which was "an experiment in on-the-fly gestural machine learning, which utilized Rebecca Fiebrink's [Wekinator](http://www.wekinator.org/) to translate motion into music." Many years later I briefly fell down another "AI art" rabbit hole with neural style transfer and guided diffusion, here are some fun results: [https://ineffable.vision/](https://ineffable.vision/) ### Advice Under Uncertainty - URL: https://raw.works/advice-under-uncertainty/ - Source: content/works/adviceunderuncertainty.md Front matter: ```json { "author": "Raymond Weitekamp", "date": "2023-09-19T00:00:00Z", "lastmod": "2023-09-19T00:00:00Z", "publishdate": "2023-09-19T00:00:00Z", "tags": [ "buffer" ], "title": "Advice Under Uncertainty" } ``` Content: {{< blockquote >}} *It is risky to offer advice for what someone else “should do”.* *It is generous to offer a “what if”, a beautiful future of the possibilities if they __chose__ to.* *It is noble to offer your vulnerable truth about the lessons you've learned and the choices you've made.* {{< /blockquote >}} It is natural to give advice. It feels like we are giving back. But is easy to overlook the complexity of the situation. We project onto our advisees. We extrapolate our lessons onto them. We are quick to forget the [uncertainty of other minds](https://plato.stanford.edu/entries/other-minds/). People are complex systems. {{< blockquote author="Nassim Taleb" link="https://www.amazon.com/Antifragile-Things-That-Disorder-Incerto-ebook/dp/B0083DJWGO?crid=J735YKQ2NE62&keywords=antifragile&qid=1695168011&sprefix=antifragile%2Caps%2C1381&sr=8-1&linkCode=ll1&tag=rawwerks09-20&linkId=5e8f3dc1f0f2c66011e441f8622219f4&language=en_US&ref_=as_li_ss_tl" title="Antifragile" >}} *Never ask anyone for their opinion, forecast, or recommendation. Just ask them what they have—or don’t have—in their portfolio.* {{< /blockquote >}} I'm trying hard to give less advice. Especially to people that I don't know well. It's easy to ["cross the net"](https://www.amazon.com/Nonviolent-Communication-Language-Life-Changing-Relationships/dp/189200528X?crid=12Q75MSL4AV3G&keywords=non-violent+communication&qid=1695168703&sprefix=non-vio%2Caps%2C241&sr=8-1&linkCode=ll1&tag=rawwerks09-20&linkId=53865d24a81a41d407d4f4212ba3a34a&language=en_US&ref_=as_li_ss_tl). It is easy to forget that [advice expires](https://world.hey.com/jason/advice-expires-d37374e6). The better you know the advisee, the better you understand the situation - the lower the risk of the advice. But the risk is never gone. For the advised, I offer you this clarifying question: *"What is the 'business model' of the person giving me this advice?".* (In the most philosophical sense of the phrase 'business model'...how do they stand to gain or lose from this advice? Do they have [skin in the game](https://www.amazon.com/s?k=skin+in+the+game+by+nassim+nicholas+taleb&crid=3K284YPMM2D8Y&sprefix=skin+in+the+game%2Caps%2C284&linkCode=ll2&tag=rawwerks09-20&linkId=4ffac524d272cf7f717311f08317e2b6&language=en_US&ref_=as_li_ss_tl)?) For now - I'm trying to focus on sharing what has worked for me, sharing my unique perspective on the situation. Who am I to say what you should do? *P.S. - I'm trying to stitch together an [Adlerian](https://www.amazon.com/Courage-Be-Disliked-Phenomenon-Happiness-ebook/dp/B078MDSV8T?crid=2QCBHT7U3DJTY&keywords=courage+to+be+disliked&qid=1695168106&sprefix=courage+%2Caps%2C219&sr=8-1&linkCode=ll1&tag=rawwerks09-20&linkId=dd870d25d1cc46a7df4cb0ac206cba1e&language=en_US&ref_=as_li_ss_tl) approach to ["skin in the game"](https://www.amazon.com/s?k=skin+in+the+game+by+nassim+nicholas+taleb&crid=3K284YPMM2D8Y&sprefix=skin+in+the+game%2Caps%2C284&linkCode=ll2&tag=rawwerks09-20&linkId=4ffac524d272cf7f717311f08317e2b6&language=en_US&ref_=as_li_ss_tl). I'm not sure where I'm headed, but I'm pretty sure that it's connected to [non-violent communication](https://www.amazon.com/Nonviolent-Communication-Language-Life-Changing-Relationships/dp/189200528X?crid=12Q75MSL4AV3G&keywords=non-violent+communication&qid=1695168703&sprefix=non-vio%2Caps%2C241&sr=8-1&linkCode=ll1&tag=rawwerks09-20&linkId=53865d24a81a41d407d4f4212ba3a34a&language=en_US&ref_=as_li_ss_tl).* ### Justifying Our Own Existence - URL: https://raw.works/justifying-our-own-existence/ - Source: content/works/justifyingourownexistence.md Front matter: ```json { "date": "2023-09-18T01:00:00-07:00", "lastmod": "2023-09-18T01:00:00-07:00", "publishdate": "2023-09-18T01:00:00-07:00", "tags": [ "LI" ], "title": "Justifying Our Own Existence" } ``` Content: After his brief internship at Twitter (now X), George Hotz concluded that the company could be run by 50 people. I'm not an expert in digital infrastructure, but 50 people doesn't sound that ridiculous to me, at least for running the product day-to-day. Maybe he's being dramatic, but the point is that the product would be functionally equivalent with 10-100x fewer people working at the company. It is human nature to protect ourselves. It is human nature to want our tribe to grow. In the case of Twitter's codebase, it seemed that there were some unintended consequences of their incentive structure... {{< blockquote author="George Hotz" link="https://lexfridman.com/george-hotz-3/" >}} *The way that you got promoted to Twitter was you wrote a library that a lot of people used, right? So some guy wrote an Nginx replacement for Twitter. Why does Twitter need an Nginx replacement? What was wrong with Nginx? Well, you see, you’re not going to get promoted if you use Nginx. But if you write a replacement and lots of people start using it as the Twitter front end for their product, then you’re going to get promoted.* {{< /blockquote >}} You get what you incentivize. Want to incentivize growth? Growth as measured by what? Headcount? It is inevitable that the inertia of the organization will eventually slow everything down. The amount of energy required to maintain the internal (human) processes becomes large, perhaps larger than the energy required to maintain the external products and processes. Bureaucracy is an emergent property. (See [Safi Bahcall's "Loon Shots"](https://amzn.to/3RIZMPx).) Within an organization, the point is not to eliminate jobs at all costs, the point is to have as many meaningful jobs as are required to run the organization. At the scale of a society - yes, we want jobs. But more importantly we want fulfilling jobs at sustainable businesses. So what are some incentive structures that can counter this tendency towards feudalism? Some ways to make it culturally honorable to replace yourself? To make a different choice than growing headcount? In terms of incentives within an organization, here are a few thoughts: - Ownership, especially if the value is much higher than the salary, and there is liquidity, or at least no penalty if you are no longer an employee. - Profit sharing. - Bonuses for solving problems without hiring additional people. In terms of market- or societal-level incentives, I'll need to keep thinking... ### Automating Social Media Posts for Hugo Sites - URL: https://raw.works/automating-social-media-posts-for-hugo-sites/ - Source: content/works/automatingsocialforhugo/index.md Front matter: ```json { "date": "2023-09-17T01:00:00-07:00", "lastmod": "2023-09-17T01:00:00-07:00", "publishdate": "2023-09-17T01:00:00-07:00", "tags": [ "LI" ], "title": "Automating Social Media Posts for Hugo Sites" } ``` Content: {{< blockquote author="Bill Gates" >}} *The first rule of any technology used in a business is that automation applied to an efficient operation will magnify the efficiency. The second is that automation applied to an inefficient operation will magnify the inefficiency.* {{< /blockquote >}} **TL;DR**: You can automatically create social media posts from specific Hugo blog posts using tags, RSS, and either [IFTTT](https://ifttt.com/applets/G7jpfrvD-rss-feed-to-linkedin) or [Make.com](https://www.make.com/en/register?pc=rawworks). Just point the automation to the tag-specific RSS link: `https://yourwebsite.com/tags/yourtag/index.xml` ![IFTTT setup](/automating-social-media-posts-for-hugo-sites/IFTTT_RSS_Hugo_to_LinkedIn.png) Once you know this trick, you can set it up in one click. Keep reading for the long boring saga of how I figured that out. This seemed really easy. Like 1 click easy. Part of my hatred for web development is that there are so many of these types of workflows that should be really easy, but aren’t because of some developer’s arbitrary decision, or because of their employer's business model. I was hoping to find a quick and easy way to automatically share my Hugo blog posts on social media. Hugo already builds RSS feed and post summaries automatically. How hard can it be to tweet those? Or post them on LinkedIn? I thought I could do it for free, in one click, with IFTTT. I could using [this IFTTT recipe](https://ifttt.com/applets/G7jpfrvD-rss-feed-to-linkedin), but then realized that this would post every single Hugo blog post to my LI profile. I wanted control over which blog posts would become social posts. IFTTT can filter by keyword, but they don’t document how the logic works and I don’t want a keyword query - I want a reliable logical filter. IFTTT can’t scrape sites and CloudFlare blocks their bots anyways. I separately looked into using either GitHub Actions or CloudFlare Workers, but they seemed too tricky for what should be a simple workflow. (Remember: I hate webdev) Then I thought: maybe I’m not finding anything because I use DuckDuckGo. I tried searching again for a solution with Google and found [Make.com](https://www.make.com/en/register?pc=rawworks) (formerly Integromat). Despite vaguely remembering that I hated Integromat, I was willing to give it a try. Unlike IFTTT, [Make.com](https://www.make.com/en/register?pc=rawworks) shows you what is going on inside the query/IF action. Now I could see the content it was scraping from RSS. I realized I could pass a tag to Hugo’s RSS template to trigger social sharing. For example the tag "LI" would mean I want to repost on LinkedIn. This was an important clue. Hugo automatically generates RSS feeds for tag pages. After some trial and error, I figured out how to target the RSS feed for a specific tag. `https://yourwebsite.com/tags/yourtag/index.xml` Or, my very explicit working example: `https://raw.works/tags/li/index.xml` Ultimately, after a few hours of fumbling around, I got automated social sharing working in one click. Use [this recipe](https://ifttt.com/applets/G7jpfrvD-rss-feed-to-linkedin) to do it for free with IFTTT. Here's a [RSS to LinkedIn template for Make](https://www.make.com/en/templates/3485-post-new-rss-items-to-linkedin) as well. This experience got me excited to explore [Make.com](https://www.make.com/en/register?pc=rawworks) more - it looks like they’ve improved the UX since Integromat and the pricing is reasonable. ### Quantity Over Quality - URL: https://raw.works/quantity-over-quality/ - Source: content/works/quantityoverquality.md Front matter: ```json { "date": "2023-09-16T01:00:00-07:00", "lastmod": "2023-09-16T01:00:00-07:00", "publishdate": "2023-09-16T01:00:00-07:00", "tags": [ "LI" ], "title": "Quantity Over Quality" } ``` Content: It is a struggle to ship work that matters. It is very easy to get in your own way, to procrastinate, to delay. It is natural to wait for the perfect phrasing, or the extra feature, or the right person to come along. The problem is that later becomes never. The recording will never sound like it did in your head. The perfect wording for you might not be the perfect wording for me. Tomorrow you might have changed your mind. Quantity, or perhaps *frequency*, is potentially a more reliable route to quality. Here is the parable excerpt from ["Art & Fear"](https://amzn.to/3LOJq47) (which is also featured in a modified form in ["Atomic Habits"](https://amzn.to/3ZrLW5S)): {{< blockquote author="Orland and Bayles" link="https://amzn.to/3LOJq47" title="Art & Fear" >}} *The ceramics teacher announced on opening day that he was dividing the class into two groups. All those on the left side of the studio, he said, would be graded solely on the "quantity" of work they produced, all those on the right solely on its "quality".* *His procedure was simple: on the final day of class he would bring in his bathroom scales and weigh the work of the “quantity” group: fifty pound of pots rated an “A”, forty pounds a “B”, and so on. Those being graded on “quality”, however, needed to produce only one pot — albeit a perfect one — to get an “A”. Well, came grading time and a curious fact emerged: the works of highest quality were all produced by the group being graded for quantity. It seems that while the “quantity” group was busily churning out piles of work – and learning from their mistakes — the “quality” group had sat theorizing about perfection, and in the end had little more to show for their efforts than grandiose theories and a pile of dead clay.* {{< /blockquote >}} What's the difference? x is not f(x). Theory is not practice. The story is not the reality. Shipping quantity forces repeated collisions with reality. Shipping frequently accelerates the velocity of learning cycles. More [OODA loops](https://en.wikipedia.org/wiki/OODA_loop) are completed. The habit of shipping regularly, of not breaking the chain, of keeping the streak going - this builds momentum. Lower the barrier to keep it fun. It doesn't need to be so serious, it just needs to mean something. It just needs to mean something to you. Remember: x is not f(x). Get addicted to the metric most in your locus of control. 10,000 steps a day is highly within your control, losing 10 lbs is less so. Publishing a meaningful post a day is highly within your control, getting 10,000 likes is less so. Let the dopamine come from the commit/send/post/publish button, not from refreshing the page while praying for shares and likes. ### Venture Games - URL: https://raw.works/venture-games/ - Source: content/works/venturegames.md Front matter: ```json { "date": "2023-09-15T01:00:00-07:00", "lastmod": "2023-09-15T01:00:00-07:00", "publishdate": "2023-09-15T01:00:00-07:00", "title": "Venture Games" } ``` Content: I used to be jealous of my friends who had raised more VC money than me. Now I am grateful to have raised less than most of my peers. Just when I felt that I was finally starting to understand the "VC game" - the entire playing field was re-arranged. A few things always bugged me about venture capital. The first was that the VCs get a portfolio of bets, but I only get one. (see: ["Ergodic Entrepreneurship"](http://rawwerks.com/ergodic-entrepreneurship/)). The second was that VCs were getting rich on companies that were absolute garbage (as long as they could be marketed properly). (ie - someone making their wealth on a company that IPO'd and then went bankrupt 18 months later, or a company that was acquired only to be shelved after discovering the "damaged goods".) I have mostly overcome these concerns. I appreciate them as part of "the venture game". In the end, you can't fool all of the people all the time. Value is perceived. Some people get lucky. Many of these problems can be addressed with [skin in the game](https://amzn.to/46hRUsd). What is happening now is that the entire market context has changed. Most of the people playing the venture game don't have any experience operating in this type of environment. I certainly don't. Most of the entrepreneurs and investors that I know were still in school during the first "cleantech bubble" burst of 2008. Almost none of them are old enough to have weathered the "dotcom bubble" of 2000. The [smart ones can do the math](https://www.linkedin.com/posts/ianrountree_vcs-look-at-this-table-look-at-it-if-you-activity-7085672045031682048-5y-4), and it is scary: {{< render-html >}} {{< /render-html >}} I don't envy VCs right now. It is a tough game to play. It is not my intention to focus on the "entrepreneurs vs. investors" aspects of the situation. While there are certainly many investors who are using the current market conditions to get good deals, that's kind of their job. As usual, Nassim Taleb captures all of these dynamics better than I can. As he [recently shared on The Tim Ferriss Show](https://tim.blog/2023/09/07/nassim-nicholas-taleb-scott-patterson/): *"Companies [were] operating on a following modus...go to the market as a cash machine, so [they] don’t even have to generate cash. [I]t has Ponzi characteristics. Someone else will buy our company or we’re packaging a company to sell it to someone else. Now, that started before the great financial crisis, but it was very moderate. Of course it took place during the crazy period of the internet bubble and then died. So we had had episodes of that effect, but now it’s ingrained. And people had now for 15 years of low interest rates. You have people in their forties who’ve never seen interest rates. And they don’t know how to behave, they don’t know how to invest. So I think the most fragile part today is not the banks of course, as we said. And it’s not hedge fund because it’s sort of like mature adults typically. It is those startups and the VCs, the venture capitalists. Venture capitalists actually played quite a nasty game because they cashed out. All of them are rich on companies that never made a penny. You see? I know there’s a lot of — take how many billionaires you have from Silicon Valley, [whose companies] never made a penny."* For myself, the biggest learning has been to focus on cash flow. This sounds like "Business 101" - and it is - it's just that my entire entrepreneurial career to date has been in an environment that has mostly incentivized growth at all costs. Now all of that has changed, and it is an important opportunity for entrepreneurs and investors alike to take the time to reflect on the new rules of the game. In my own business, I've been trying to heed [Jason Friedman's advice](https://world.hey.com/jason/you-only-compete-with-one-thing-48a20d93): *"Of course you compete in a market with multiple options, but that's not your real competition. Your real competition are your costs. Yours, not theirs. You, not them."* ### There Will Be Blood - Pt. 1 - URL: https://raw.works/there-will-be-blood-pt.-1/ - Source: content/works/therewillbeblood-pt1.md Front matter: ```json { "date": "2023-09-14T01:00:00-07:00", "lastmod": "2023-09-14T01:00:00-07:00", "publishdate": "2023-09-14T01:00:00-07:00", "title": "There Will Be Blood - Pt. 1" } ``` Content: The date of the surgery was set months in advance, but not the time or location. A few days in advance, a call was promised to confirm the time and location - but no one called. No one picked up the phone when they were called, even though there were two separate extension lines, with separate voicemails, one of which was full. No one picked up the phone during the lunch break, and of course messages cannot be left during the lunch break. I am desperate for details and less than 24 hours from the surgery. The first person to pick up the phone got the time correct, but the location wrong. Finally, someone called the night before the surgery, but the phone number was not the same area code as the hospital. This person got the location right, but the time wrong. The pre-op instructions given verbally explicitly contradicted the pre-op package that was sent to the house. *"Drink the [ClearFast](https://amzn.to/44VbYzr) at 3am." "Really?" "Well maybe midnight would be okay, but don't drink it within 4 hours of arrival."* (The bottle says to drink exactly 2 hours before arrival) *"Can I have black coffee?" "No."* (The pre-op checklist says yes) While the surgery is already underway, someone sends a "secure message" responding to my voicemail, hoping that I go to the right location in time. (But gets the time wrong.) The one detail that everyone got correct? There will be blood. ### Cosmic Reassurance - URL: https://raw.works/cosmic-reassurance/ - Source: content/works/fearnotloss.md Front matter: ```json { "date": "2023-09-13T01:00:00-07:00", "lastmod": "2023-09-13T01:00:00-07:00", "publishdate": "2023-09-13T01:00:00-07:00", "title": "Cosmic Reassurance" } ``` Content: Fear not loss. Ruin rerises reformed. ### Xanadu Now - URL: https://raw.works/xanadu-now/ - Source: content/works/xanadu now.md Front matter: ```json { "date": "2023-09-12T01:00:00-07:00", "lastmod": "2023-09-12T01:00:00-07:00", "publishdate": "2023-09-12T01:00:00-07:00", "title": "Xanadu Now" } ``` Content: Sleep in Xanadu's Pleasure Dome - One Night Stay Includes Guided River Tour Wander Kubla Khan's Lush Gardens and Ancient Forests - Immersive 2-Day Escape Enjoy an Exotic Concert by the Abyssinian Maid - Private Dulcimer Performance and Feast Experience Kubla's Erupting Fountain First-hand - Behind the Scenes Geological Tour Sleep Under the River Alph - Overnight Stay in the Caverns Measureless to Man Monthly Milk of Paradise Subscription - Get Xanadu's Divine Nectar Shipped to Your Door [*Or, a vision in a dream. A Fragment.*](https://amzn.to/44OenvJ) ### Paper in Your Pocket - URL: https://raw.works/paper-in-your-pocket/ - Source: content/works/Paper in Your Pocket.md Front matter: ```json { "date": "2023-09-11T01:00:00-07:00", "lastmod": "2023-09-11T01:00:00-07:00", "publishdate": "2023-09-11T01:00:00-07:00", "title": "Paper in Your Pocket" } ``` Content: The insight scribbled hastily On a scrap of paper in your pocket: How will it reach others? Like a single pebble cast into still waters, Its ripples travel far. Yet if shared in haste It becomes just another rock Lost in the cairn. Lower the barrier Smooth the way Make space for the unspoken. In the beginner’s mind Scattered seeds become flowers Frustration melts like morning frost Immediate praise and blame Are dancing shadows. The one who shares, the one who hears- Not two. Like the moon in the lake Insight reflects insight. The interface is empty, yet fulfilled. Let your insight fall like rain Seep through cracks to sow its seed. A quiet truth echoes loudest. ### On Feedback - URL: https://raw.works/on-feedback/ - Source: content/works/onfeedback.md Front matter: ```json { "date": "2023-09-10T01:00:00-07:00", "lastmod": "2023-09-10T01:00:00-07:00", "publishdate": "2023-09-10T01:00:00-07:00", "title": "On Feedback" } ``` Content: How long can you write if no one is reading? How much can you work if no one is paying? As much as we want to see ourselves as steering our own ship, following our own nose, forging our own path — most of us are not Van Gogh. We need feedback and recognition to thrive. Just enough. Too little = Van Gogh. Too much = arrested development. Enough to know we’re not alone. Not so much that we can’t hear our own voice. ### Blue Moon - URL: https://raw.works/blue-moon/ - Source: content/works/bluemoon.md Front matter: ```json { "date": "2023-09-09T01:00:00-07:00", "lastmod": "2023-09-09T01:00:00-07:00", "publishdate": "2023-09-09T01:00:00-07:00", "title": "Blue Moon" } ``` Content: The red sun rises in the rearview, Blazing bright over the highway behind me. The blue moon sets ahead over stagnant cars, Sinking below the foggy ridgeline. The Dumbarton smells like farts. For a moment I'm transported, certain I'm there - On the playa, breathing it all in. In the morning twilight, farts and dust. ### Understanding Stories - URL: https://raw.works/understanding-stories/ - Source: content/works/understandingstories.md Front matter: ```json { "date": "2023-09-08T01:00:00-07:00", "lastmod": "2023-09-08T01:00:00-07:00", "publishdate": "2023-09-08T01:00:00-07:00", "title": "Understanding Stories" } ``` Content: There is a fear of AI writing stories for us, as Harari says ["because we don’t understand what’s behind it"](https://youtu.be/Mde2q7GFCrw?si=MgSoRHPAwcomln-K&t=7452). But I’m not sure we’ve ever understood [where our stories come from](/where-do-stories-come-from/). I’ve written a handful of songs. I could probably tell you what they are about, but I would be lying if I said that I knew where they came from. (Here's the full interview, below) {{< youtube Mde2q7GFCrw >}} ### Chief No Officer - URL: https://raw.works/chief-no-officer/ - Source: content/works/chiefnoofficer.md Front matter: ```json { "date": "2023-09-07T01:00:00-07:00", "lastmod": "2023-09-07T01:00:00-07:00", "publishdate": "2023-09-07T01:00:00-07:00", "tags": [ "LI" ], "title": "Chief No Officer" } ``` Content: I met a C-level leader at a 3D printing startup that described his functional role as the “Chief No Officer”. His job was to say no to almost everything. The promise of 3D printing is that it can it can (supposedly) make everything. The reality is that it can rarely make anything. His version of this was "just because you can do something doesn't mean you should do it". Hence the CNO's job was to shut down most of the ideas, to keep the company focused. Reflecting on this interaction, I am seeing how I have not been very good at saying no. The potentiality of yes is exciting. The possibilities. The novelty. The excitement. The reality of yes is exhausting. Missed deadlines. Unresponded-to emails. Disappointed customers and partners. It is hard to say no. It takes [the courage to be disliked](https://amzn.to/45ExNVa). We can never please everyone. Adler explained the impossibility of this. Without the discipline of no, it is hard to have time to focus on the few truly resonant yeses. Moving forward, I'm going to try to be a better CNO, to apply the ["Hell Yeah, or No"](https://amzn.to/3ELr7ZH) Rule a bit more often. ### Mixed Messages - URL: https://raw.works/mixed-messages/ - Source: content/works/mixedmessages/index.md Front matter: ```json { "date": "2023-09-06T01:00:00-07:00", "lastmod": "2023-09-06T01:00:00-07:00", "publishdate": "2023-09-06T01:00:00-07:00", "title": "Mixed Messages" } ``` Content: ![Are you sending mixed messages?](/mixed-messages/mixedmessages.jpg) Are you sending mixed messages? ### Not planned, intended - URL: https://raw.works/not-planned-intended/ - Source: content/works/notplannedintended.md Front matter: ```json { "date": "2023-09-05T01:00:00-07:00", "lastmod": "2023-09-05T01:00:00-07:00", "publishdate": "2023-09-05T01:00:00-07:00", "title": "Not planned, intended" } ``` Content: The best things that have happened in my life were not planned, they were intended. I did not plan to meet my wife. I did have an intention of being open to a romantic relationship. I did not plan to join Cyclotron Road. I did have an intention to combine my interests in physical science & entrepreneurship, as well as an intention to live in the Bay Area. Are you making a plan, or setting an intention? ### Multiple Choice: Moments - URL: https://raw.works/multiple-choice-moments/ - Source: content/works/momentsmultiplechoice.md Front matter: ```json { "date": "2023-09-04T01:00:00-07:00", "lastmod": "2023-09-04T01:00:00-07:00", "publishdate": "2023-09-04T01:00:00-07:00", "title": "Multiple Choice: Moments" } ``` Content: Multiple Choice Monday. In my in-between moments, I am most likely to: - Scroll - Text - Tap - Type - Feel - Breathe - Ask - Swipe - Sense - Sit - Pee - Wipe - Worry - Rest - Rage - Snipe - (None of the above) ### AI doesn't blink - URL: https://raw.works/ai-doesnt-blink/ - Source: content/works/AIdoesntblink.md Front matter: ```json { "date": "2023-09-03T01:00:00-07:00", "lastmod": "2023-09-03T01:00:00-07:00", "publishdate": "2023-09-03T01:00:00-07:00", "title": "AI doesn't blink" } ``` Content: [A.I.](https://amzn.to/3Z1ffvV) doesn’t blink. This was the vision of Haley Joel Osment and Steven Spielberg. The [red doorknob](https://amzn.to/47SYkzB). The [origami unicorn](https://amzn.to/47Tet8a). The subtle tell that is only obvious when it’s too late. Make sure your friends are blinking. ### Where do stories come from? - URL: https://raw.works/where-do-stories-come-from/ - Source: content/works/wheredostoriescomefrom.md Front matter: ```json { "date": "2023-09-02T01:00:00-07:00", "lastmod": "2023-09-02T01:00:00-07:00", "publishdate": "2023-09-02T01:00:00-07:00", "title": "Where do stories come from?" } ``` Content: My friend manifests stories. She doesn’t write them, she writes them down. She has a protocol. She asks for them. From whom? I’m not sure she could tell you exactly, even if she wanted to. Mommy, where do stories come from? ### Superpowers - URL: https://raw.works/superpowers/ - Source: content/works/superpowers.md Front matter: ```json { "date": "2023-09-01T01:00:00-07:00", "lastmod": "2023-09-01T01:00:00-07:00", "publishdate": "2023-09-01T01:00:00-07:00", "title": "Superpowers" } ``` Content: The very tools that promise to give us superpowers simultaneously (and perhaps necessarily) erode our intrinsic superpowers. If we haven’t yet discovered our innate superpowers, these technologies make it very hard to get quiet enough to notice them. Especially when our tools talk back, they can make it difficult to find time enough (and attention enough) to hone our (super)natural (super)powers. ### A child's nightmare - URL: https://raw.works/a-childs-nightmare/ - Source: content/works/childsnightmare.md Front matter: ```json { "date": "2023-08-31T01:00:00-07:00", "lastmod": "2023-08-31T01:00:00-07:00", "publishdate": "2023-08-31T01:00:00-07:00", "title": "A child's nightmare" } ``` Content: As a small child, I had a recurring nightmare of getting trapped inside of the family computer. It took me more than 30 years to appreciate the wisdom in this dream. It was not the irrational fear of a four year old boy. It was a warning, a prophesy -- and it has now come true. ### Job with a capital J - URL: https://raw.works/job-with-a-capital-j/ - Source: content/works/jobwithacapitalj.md Front matter: ```json { "date": "2023-08-30T01:00:00-07:00", "lastmod": "2023-08-30T01:00:00-07:00", "publishdate": "2023-08-30T01:00:00-07:00", "title": "Job with a capital J" } ``` Content: What is your Job? (with a capital J) A story more powerful than money. A calling. A vocation. Where did it come from? How did you find it? What is in the way of you doing your Job? ### Garbage Truck Mantra - URL: https://raw.works/garbage-truck-mantra/ - Source: content/works/garbagetrucks.md Front matter: ```json { "date": "2023-08-29T01:00:00-07:00", "lastmod": "2023-08-29T01:00:00-07:00", "publishdate": "2023-08-29T01:00:00-07:00", "title": "Garbage Truck Mantra" } ``` Content: Some days, no matter where you walk, you can’t seem to avoid the garbage trucks ### Whose problem? - URL: https://raw.works/whose-problem/ - Source: content/works/whoseproblem.md Front matter: ```json { "date": "2023-08-28T01:00:00-07:00", "lastmod": "2023-08-28T01:00:00-07:00", "publishdate": "2023-08-28T01:00:00-07:00", "title": "Whose problem?" } ``` Content: When someone tells you, *"If you don’t do \___, that’s gonna be a problem,"* they probably have a problem that they are not telling you about. They see you as the way to solve their problem, even if only psychologically. ### Bliss & Rent - URL: https://raw.works/bliss-rent/ - Source: content/works/bliss and rent.md Front matter: ```json { "date": "2023-08-27T01:00:00-07:00", "lastmod": "2023-08-27T01:00:00-07:00", "publishdate": "2023-08-27T01:00:00-07:00", "title": "Bliss \u0026 Rent" } ``` Content: Find your bliss, and pay your rent. *...always remember...* Most landlords do not accept bliss for payment, and true bliss is not for sale. ### Quotes - URL: https://raw.works/quotes/ - Source: content/quotes.md Front matter: ```json { "layout": "quotes", "quote_list": [ { "author": "Steve Jobs", "index": "1", "quote": "Focusing is about saying No.", "slug": "steve-jobs-focus-saying-no" }, { "author": "Mahatma Gandhi", "index": "2", "quote": "Whenever you are confronted with an opponent, conquer them with love.", "slug": "mahatma-gandhi-conquer-love" }, { "author": "Maya Angelou", "index": "3", "quote": "You may not control all the events that happen to you, but you can decide not to be reduced by them.", "slug": "maya-angelou-decide-strength" }, { "author": "Seth Godin", "index": "4", "quote": "Most people I care about don't want to be hustled", "slug": "seth-godin-hustled-unwanted" }, { "author": "Seth Godin", "index": "5", "quote": "you are not going to be able to interrupt your way into success", "slug": "seth-godin-interrupt-fail" }, { "author": "Seth Godin", "index": "6", "quote": "would they miss you if you didn't show up?", "slug": "seth-godin-miss-show-matters" }, { "author": "seth godin", "index": "7", "quote": "it might not be for you, but if it is, here i am", "slug": "seth-godin-here-purpose" }, { "author": "seth godin", "index": "8", "quote": "if someone else decides it is worth talking about, it is remarkable", "slug": "seth-godin-worth-talking-remarkable" }, { "author": "Sven Goran Eriksson", "index": "9", "quote": "The greatest barrier to success is the fear of failure.", "slug": "sven-goran-eriksson-fear-failure-barrier" }, { "author": "Seth Godin", "index": "10", "quote": "That thing you are making, does my life get better if I tell other people about it?", "slug": "seth-godin-better-telling" }, { "author": "Eleanor Roosevelt", "index": "11", "quote": "The future belongs to those who believe in the beauty of their dreams.", "slug": "eleanor-roosevelt-dreams-determine-future" }, { "author": "Hermann Hesse", "index": "12", "quote": "Within you is a stillness and a sanctuary to which you can retreat at any time and be yourself.", "slug": "hermann-hesse-sanctuary-within" }, { "author": "René Descartes", "index": "13", "quote": "Divide each difficulty into as many parts as is feasible and necessary to resolve it.", "slug": "rene-descartes-divide-resolve" }, { "author": "Billy Cox", "index": "14", "quote": "The more you think and talk about your goals, the more positive and enthusiastic you become.", "slug": "billy-cox-enthusiastic-goals" }, { "author": "Buddha", "index": "15", "quote": "Your worst enemy cannot harm you as much as your own unguarded thoughts.", "slug": "buddha-thoughts-harm" }, { "author": "Dalai Lama", "index": "16", "quote": "Be kind whenever possible. It is always possible.", "slug": "dalai-lama-always-kind-possible" }, { "author": "Eleanor Roosevelt", "index": "17", "quote": "No one can make you feel inferior without your consent.", "slug": "eleanor-roosevelt-consent-inferior" }, { "author": "Maya Angelou", "index": "18", "quote": "All great achievements require time.", "slug": "maya-angelou-time-achievement" }, { "author": "Dean Acheson", "index": "19", "quote": "Always remember that the future comes one day at a time.", "slug": "dean-acheson-future-day-time" }, { "author": "Voltaire", "index": "20", "quote": "The longer we dwell on our misfortunes, the greater is their power to harm us.", "slug": "voltaire-harm-dwell" }, { "author": "Earl Nightingale", "index": "21", "quote": "Our attitude toward life determines life's attitude towards us.", "slug": "earl-nightingale-attitude-determines" }, { "author": "Galileo Galilei", "index": "22", "quote": "We cannot teach people anything. We can only help them discover it within themselves.", "slug": "galileo-galilei-discover-within" }, { "author": "Bruce Lee", "index": "23", "quote": "Always be yourself, express yourself, have faith in yourself. Do not go out and look for a successful personality and duplicate it.", "slug": "bruce-lee-yourself-unique" }, { "author": "Denis Waitley", "index": "24", "quote": "There are two primary choices in life: to accept conditions as they exist, or accept the responsibility for changing them.", "slug": "denis-waitley-change-accept" }, { "author": "Bruce Lee", "index": "25", "quote": "Long-term consistency trumps short-term intensity.", "slug": "bruce-lee-consistency-intensity" }, { "author": "Malcolm Forbes", "index": "26", "quote": "Failure is success if we learn from it.", "slug": "malcolm-forbes-learn-failure" }, { "author": "Eleanor Roosevelt", "index": "27", "quote": "The future belongs to those who believe in the beauty of their dreams.", "slug": "eleanor-roosevelt-future-beauty-dreams" }, { "author": "Eric Hoffer", "index": "28", "quote": "In times of change, learners inherit the earth, while the learned find themselves beautifully equipped to deal with a world that no longer exists.", "slug": "eric-hoffer-change-learn-adapt" }, { "author": "Norwegian proverb", "index": "29", "quote": "A hero is one who knows how to hang on for one minute longer.", "slug": "norwegian-proverb-hero-hang" }, { "author": "Abraham Maslow", "index": "30", "quote": "In any given moment we have two options: to step forward into growth or back into safety.", "slug": "abraham-maslow-growth-safety" }, { "author": "Steve Jobs", "index": "31", "quote": "Sometimes life hits you in the head with a brick. Don't lose faith.", "slug": "steve-jobs-faith-brick" }, { "author": "Jerry West", "index": "32", "quote": "You can’t get much done in life if you only work on the days when you feel good.", "slug": "jerry-west-work-feel" }, { "author": "Thucydides", "index": "33", "quote": "The secret to happiness is freedom. And the secret to freedom is courage.", "slug": "thucydides-freedom-courage" }, { "author": "Ralph Waldo Emerson", "index": "34", "quote": "Unless you try to do something beyond what you have already mastered you will never grow.", "slug": "ralph-waldo-emerson-grow-master" }, { "author": "Harriet Braiker", "index": "35", "quote": "Striving for excellence motivates you; striving for perfection is demoralizing.", "slug": "harriet-braiker-excellence-perfection" }, { "author": "Dr Roopleen", "index": "36", "quote": "If you have a dream, don’t just sit there. Gather courage to believe that you can succeed and leave no stone unturned to make it a reality.", "slug": "dr-roopleen-succeed-reality" }, { "author": "Abraham Hicks", "index": "37", "quote": "If you are not excited about it, it may not be the right path.", "slug": "abraham-hicks-excited-rig" }, { "author": "Dalai Lama", "index": "38", "quote": "Remember that not getting what you want is sometimes a wonderful stroke of luck.", "slug": "dalai-lama-remember-not-getting-wonderful" }, { "author": "James Joyce", "index": "39", "quote": "Mistakes are the portals of discovery.", "slug": "james-joyce-mistakes-discovery" }, { "author": "Gabby Bernstein", "index": "40", "quote": "Give yourself permission to slow down. You can speed up by slowing down.", "slug": "gabby-bernstein-give-slow-speed" }, { "author": "Mae Jemison", "index": "41", "quote": "Never be limited by other people’s limited imaginations.", "slug": "mae-jemison-never-limited-imaginations" }, { "author": "Henry David Thoreau", "index": "42", "quote": "This world is but canvas to our imaginations.", "slug": "henry-david-thoreau-world-canvas-imaginations" }, { "author": "Marcel Proust", "index": "43", "quote": "If a little dreaming is dangerous, the cure for it is not to dream less but to dream more, to dream all the time.", "slug": "marcel-proust-dream-dangerous-cure-dream" }, { "author": "Ralph Waldo Emerson", "index": "44", "quote": "Don’t be pushed by your problems. Be led by your dreams.", "slug": "ralph-waldo-emerson-problems-dreams" }, { "author": "Alan Cohen", "index": "45", "quote": "Don’t postpone joy until you have learned all of your lessons. Joy is your lesson.", "slug": "alan-cohen-dont-postpone-joy-lesson" }, { "author": "Mahatma Gandhi", "index": "46", "quote": "Action expresses priorities.", "slug": "mahatma-gandhi-action-priorities" }, { "author": "Albert Einstein", "index": "47", "quote": "Creativity is contagious, so pass it on.", "slug": "albert-einstein-creativity-contagious" }, { "author": "Michelle Obama", "index": "48", "quote": "Be focused. Be determined. Be hopeful. Be empowered.", "slug": "michelle-obama-focused-determined-hopeful-empowered" }, { "author": "Robert Schuller", "index": "49", "quote": "Today's accomplishments were yesterday's impossibilities.", "slug": "robert-schuller-todays-yesterday-impossibilities" }, { "author": "John C. Maxwell", "index": "50", "quote": "The only guarantee for failure is to stop trying.", "slug": "john-c-maxwell-only-failure-stop" }, { "author": "Amelia Earhart", "index": "51", "quote": "The most effective way to do it, is to do it.", "slug": "amelia-earhart-effective-do" }, { "author": "Maya Angelou", "index": "52", "quote": "People will forget what you said, people will forget what you did, but people will never forget how you made them feel.", "slug": "maya-angelou-forget-said-did-feel" }, { "author": "Vincent Van Gogh", "index": "53", "quote": "Great things are done by a series of small things brought together.", "slug": "vincent-van-gogh-great-small-together" }, { "author": "Cindy Francis", "index": "54", "quote": "Accept life as it is. Then work to make it the way you want it to be.", "slug": "cindy-francis-accept-work-want" }, { "author": "Satsuki Shibuya", "index": "55", "quote": "By doing what you love you inspire and awaken the hearts of others.", "slug": "satsuki-shibuya-love-awaken" }, { "author": "Proverb", "index": "56", "quote": "A smooth sea never made a skillful sailor.", "slug": "proverb-smooth-skillful" }, { "author": "Fred Rogers", "index": "57", "quote": "Often when you think you're at the end of something, you're at the beginning of something else.", "slug": "fred-rogers-end-beginning" }, { "author": "Amelia Earhart", "index": "58", "quote": "Decide whether or not the goal is worth the risks involved. If it is, stop worrying.", "slug": "amelia-earhart-goal-risks-worry" }, { "author": "Archimedes, 'Sword in the Stone'", "index": "59", "quote": "The only way to learn it is to do it.", "slug": "archimedes-sword-stone-learn-do" }, { "author": "Ralph Waldo Emerson", "index": "60", "quote": "What lies behind you and what lies in front of you, pales in comparison to what lies inside of you.", "slug": "ralph-waldo-emerson-lies-inside" }, { "author": "Saadi", "index": "61", "quote": "Have patience. All things are difficult before they become easy.", "slug": "saadi-patience-difficult-easy" }, { "author": "Captain Jack Sparrow", "index": "62", "quote": "The problem is not the problem. The problem is your attitude about the problem.", "slug": "captain-jack-sparrow-attitude-problem" }, { "author": "Winston Churchill", "index": "63", "quote": "A pessimist sees the difficulty in every opportunity; an optimist sees the opportunity in every difficulty.", "slug": "winston-churchill-pessimist-optimist" }, { "author": "Hannu Rajaniemi", "index": "64", "quote": "If reality is not what you want it to be, change it.", "slug": "hannu-rajaniemi-reality-change" }, { "author": "Bruce Lee", "index": "65", "quote": "Absorb what is useful. Discard what is not. Add what is uniquely your own.", "slug": "bruce-lee-discard-add" }, { "author": "Seneca", "index": "66", "quote": "It is not because things are difficult that we do not dare; it is because we do not dare that they are difficult.", "slug": "seneca-difficult-dare" }, { "author": "Richard Branson", "index": "67", "quote": "Spend more time smiling than frowning and more time praising than criticizing.", "slug": "richard-branson-smile-praise" }, { "author": "Jonathan Safran Foer", "index": "68", "quote": "You cannot protect yourself from sadness without protecting yourself from happiness.", "slug": "jonathan-safran-foer-protect-happiness-sadness" }, { "author": "Bruce Lee", "index": "69", "quote": "I'm not in this world to live up to your expectations and you're not in this world to live up to mine.", "slug": "bruce-lee-expectations" }, { "author": "Dalai Lama", "index": "70", "quote": "When you talk, you are only repeating what you already know. But if you listen, you may learn something new.", "slug": "dalai-lama-listen" }, { "author": "80/20", "index": "71", "quote": "Become unique usefully and joyfully", "slug": "richard-branson-smile-praise" }, { "author": "Courage to be Happy", "index": "72", "quote": "All chronologies and history books are apocrypha compiled for the purpose of proving the legitimacy of those currently in power.", "slug": "courage-to-be-happy-chronology" }, { "author": "Courage to be Happy", "index": "73", "quote": "Instead of placing your worth on being different from people, place your worth on being yourself", "slug": "courage-to-be-happy-be-yourself" } ], "title": "Quotes" } ``` Content: something ### Sign up error - URL: https://raw.works/subscribe-fail/ - Source: content/subscribe-fail.md Front matter: ```json { "hidefooter": true, "hidesidebar": true, "layout": "single", "title": "Sign up error" } ``` Content: Sorry, we weren't able to sign you up. Want to try again? {{< email-signup >}} ### Thanks for signing up! - URL: https://raw.works/subscribe-success/ - Source: content/subscribe-success.md Front matter: ```json { "hidefooter": true, "layout": "single", "title": "Thanks for signing up!" } ``` Content: Thanks for signing up!