korrents

documentation

FilterEveryone, all time55 from 46 people
Group All Technologists 11Scientists 7Economists & investors 1Founders & executives 1Writers & journalists 5Philosophers 1
Who Everyone RFRichard Feynman 3 ARAlice Ryhl 2 MTMeng To 2 PMPatrick McKenzie 2 SKSteve Krug 2 SSSusan Sontag 2 TSTobias Sjösten 2 VBVicki Boykis 2 ASAlice Schroeder 1 AKAndrej Karpathy 1 ADAnil Dash 1 ARArmin Ronacher 1 BGBrendan Gregg 1 BTBrett Terpstra 1 CBCarlos Alexandro Becker 1 CMCasey Muratori 1 DKDaniel Kahneman 1 DLDerek Lowe 1 DHDex Horthy 1 FCFrançois Chollet 1 HHHamel Husain 1 HAHannah Arendt 1 HKHenrik Karlsson 1 HWHillel Wayne 1 JWJack Weatherford 1 JAJen Abel 1 JEJulia Evans 1 KHKelsey Hightower 1 KRKenneth Reitz 1 KVKenton Varda 1 LOLuis Ouriach 1 MNMark Nottingham 1 MPMaxence Poutord 1 MLMichael Lopp 1 NPNate Parrott 1 NENelson Elhage 1 PKPeter Kropotkin 1 PSPeter Steinberger 1 PWPeter Woit 1 RDRay Dalio 1 RCRuss Cox 1 SSSalvatore Sanfilippo 1 SCSasha Chapin 1 SHStephen Hawking 1 TSTara Seshan 1 YXYihui Xie 1
When All time Past 1 yearPast 3 yearsPast 5 years
  1. AR

    Armin Ronacher quoted

    Creator of Flask and Jinja

    Our readingAI agents can maintain kid-friendly interactive documentation that explains how game code works.

    AI agents

    Their wordsnot only can you build a game with kids with AI, you can also have the agent maintain kids friendly interactive documentation along the way that explains how it works, linked up to the actual game code.

    ↗@mitsuhiko on Xx.com

  2. 3 weeks earlier
  3. BT

    Brett Terpstra quoted

    Mac developer and writer at brettterpstra.com

    Our readingDifferent Markdown flavors force users to trade off features like citations, task lists, and table support against each other.

    Their wordsMarkdown flavors keep splitting. Pick MultiMarkdown for citations and you lose something else. Pick GFM for task lists and you're fighting table quirks from another tool.

    ↗Apex comes to Marked 3brett.trpstra.net

  4. 1 day earlier
  5. FC

    François Chollet quoted

    Creator of the Keras deep-learning library and the ARC-AGI benchmark

    Our readingThose who relied on code as their source of truth for managing understanding now need a new source-of-truth artifact.

    Their wordsIf code was previously your source of truth for managing understanding, then you need a new source of truth artifact.

    ↗@fchollet on Xx.com

  6. 11 days earlier
  7. HK

    Henrik Karlsson quoted

    Swedish essayist who writes Escaping Flatland

    Our readingPhotographs do not represent the world the way human perception actually experiences it.

    Their wordsThe problem with his old images, and with photography, was that that is not at all what the world looks like to us.

    ↗Some remarks on being stuck, with digressions on David Hockney, developing style, and gardenshenrikkarlsson.xyz 2nd of 3 in this piece

  8. 1 day earlier
  9. PW

    Peter Woit quoted

    Mathematician and Senior Lecturer in the Mathematics…

    Our readingMathematicians using AI agents to produce proofs should have to disclose in detail how the agents were used, the way experimental scientists must describe their methods.

    mathematicsAI agents

    Their wordsthe math community needs to adopt a version of the ethical standards of experimental science. If you are using AI agents, you can't just give a proof (formalized or not), but need to also provide a detailed explanation of how these agents were used to get the result.

    ↗A Modest Proposalmath.columbia.edu 2nd of 2 in this piece

  10. 8 days earlier
  11. SC

    Sasha Chapin quoted

    Writer and author of All the Wrong Moves

    Our readingProfound transformations of the mind are real and underappreciated, and that fact deserves to be preserved as knowledge.

    Their wordsIt's really not obvious that such profound transformations can happen to a mind, and it's really important that this knowledge gets preserved.

    ↗Biting the no-self bulletsashachapin.substack.com 2nd of 3 in this piece

  12. 2 days earlier
  13. TS

    Tara Seshan quoted

    Product lead for ChatGPT Work at OpenAI

    Our readingSomething people can try, or results from a test, communicates a direction far better than a document.

    Their wordsI have like results where we tried this, we ran an AB, here's like the results. This is why I think we should like go in this direction. That is a way better communication tool than like the the doc itself.

    ↗AI’s third era: the rise of persistent AI coworkers | Tara Seshan (OpenAI’s product lead)youtube.com 15th of 17 in this recording

  14. 7 days earlier
  15. JA

    Jen Abel quoted

    Co-founder of JJELLYFISH and GM of enterprise sales at State Affairs

    Our readingA seller should not record the intro call, because buyers will not be open on a recording.

    Their wordsDo not bring a recorder to this call. Do not record the call. They will not be open. They will not be vulnerable.

    ↗84 minutes of enterprise sales alpha | Jen Abelyoutube.com 7th of 24 in this recording

  16. 4 days earlier
  17. ML

    Michael Lopp quoted

    Engineering manager who has worked at Apple, Netscape

    Our readingThe people who would most need to disclose AI-generated content are the least motivated to actually disclose it.

    Their wordsThe content for which you want this the most would be the responsibility of humans who are the least motivated to use such a system.

    ↗How Grumbles Helpsrandsinrepose.com 1st of 3 in this piece

  19. KR

    Kenneth Reitz quoted

    Programmer, photographer and musician

    Our readingUnderstanding what a system does does not reveal what it fundamentally is.

    Their wordsFunction is not ontology. A working API specification tells me what calls are available and how the system behaves at the boundary.

    ↗The Mind Builds an APIkennethreitz.org 3rd of 3 in this piece

  20. 7 days earlier
  21. VB

    Vicki Boykis quoted

    Machine learning engineer working on recommender systems

    Our readingExplanatory text and jargon generated by machines for readers with different levels of understanding makes communication worse.

    Their wordsBut where it starts to get dicey is when you have both jargon and long text explanations being generated by machines, for people at different levels of understanding.

    ↗Write for peoplevickiboykis.com 1st of 3 in this piece

  23. VB

    Vicki Boykis quoted

    Our readingAI-generated technical documentation is optimized for completing a sentence rather than for being understood by a human reader.

    Their wordsThe output is the result of a model trained on the entire corpus of technical internet documentation with the training objective being completing a sentence but not with the objective of human understanding.

    ↗Write for peoplevickiboykis.com 2nd of 3 in this piece

  24. 8 days earlier
  25. NP

    Nate Parrott quoted

    Designer at Anthropic who started Claude Design

    Our readingWriting a doc is an uphill battle to get people to read, and the best way to sell people on ideas is a designer's mockup or prototype.

    design

    Their wordsand every PM knows like you write a doc and you know it's kind of an uphill battle to get people to read it and the best way to sell people on ideas is to get a designer to mock up or prototype the actual idea.

    ↗Nate Parrott - How a Side Project Became Claude Designyoutube.com 8th of 19 in this recording

  26. 6 days earlier
  27. HW

    Hillel Wayne quoted

    Writer and consultant on formal methods

    Our readingOther engineering fields compile deep reference books about their materials and software has nothing of the kind, not even a five-hundred-page book on how to version an API.

    Their wordsAnd that kind of compiling of information about the materials is something other fields do that we don't do. An analogy that I would think of in software would be something like a 500-page book on how to version an API.

    ↗Formal methods with Hillel Wayneyoutube.com 7th of 24 in this recording

  28. 7 days earlier
  29. SS

    Salvatore Sanfilippo quoted

    Programmer; wrote Redis and hping, and blogs about C

    Our readingDocumentation should be written for AI coding agents to understand and modify software, not only for humans.

    coding agents

    Their wordsAlso, it means that the documentation itself should not be just good for humans, but also for coding agents to understand how to change the system.

    ↗Not just development, distribution of software may change as wellantirez.com 2nd of 2 in this piece

  30. 7 days earlier
  31. DH

    Dex Horthy quoted

    Founder of the agent company HumanLayer

    Our readingNobody has found keeping specs and code in sync worth the upkeep; the code stays the source of truth.

    Their wordsI don't think anyone found it useful enough to like maintain a system to keep the specs and the code in sync versus just using the code as the source of truth always.

    ↗Context engineering with Dex Horthyyoutube.com 19th of 26 in this recording

  32. 5 days earlier
  33. YX

    Yihui Xie quoted

    Statistician and software engineer who writes R packages…

    Our readingStandardizing document or table tooling on HTML instead of PDF, LaTeX, or RTF removes most of the underlying complexity.

    Their wordsOnce you commit to HTML, an enormous amount of complexity simply evaporates.

    ↗The lt Package: Lightweight HTML Tables for R (and Beyond)yihui.org

  34. 10 days earlier
  35. MT

    Meng To quoted

    Designer and founder of Design+Code

    Our readingThe design.md is basically a designer's style guide, all the sweat, tears and blood of their mastery, in one markdown file that the AI can read.

    design

    Their wordsall the designers, they have a style guide. It's kind of like their guiding principle of their mastery of a designer. Right? You go to school, you you work on 10,000 designs, you have embedded in your blood that design language, that style guide. The design.md is basically all of that sweat, tears, and blood into one markdown file that the AI can read.

    ↗Meng To - Codex superpowers for designersyoutube.com 14th of 18 in this recording

  37. MT

    Meng To quoted

    Our readingThe design.md is powerful at the beginning, but when you are doing a lot of iterations really quickly you don't want to always reference it, and you can save it into agents.md.

    design

    Their wordsdesign.md is powerful at the beginning, but when you're doing a lot of iterations really quickly, you don't want to always reference that. You can totally save that into your agents.md.

    ↗Meng To - Codex superpowers for designersyoutube.com 16th of 18 in this recording

  38. 4 weeks earlier
  39. KH

    Kelsey Hightower quoted

    Self-taught software engineer

    Our readingIf you will write a huge markdown file so an agent has context, you should have written the documentation that gives a human the same context.

    Their wordsI'm watching people write these huge markdown files to give the agent context. How about you write documentation to give me context so I can have fully working examples.

    ↗Kubernetes and retiring at the top with Kelsey Hightoweryoutube.com 24th of 28 in this recording

  40. 8 days earlier
  41. CB

    Carlos Alexandro Becker quoted

    Brazilian software engineer

    Our readingA portfolio shows that a person's work is visible, not that they are skilled.

    Their wordsA portfolio does not make you good. It makes your work visible.

    ↗AI didn't kill portfolioscarlosbecker.com 3rd of 3 in this piece

  42. 6 days earlier
  43. AR

    Alice Ryhl quoted

    Software engineer on Google's Android Rust team

    Our readingExamples in documentation should be compiled and run as tests, so that changing the code breaks the docs instead of quietly outdating them.

    Their wordsNow the thing is you can of course write examples in your documentation and Rust makes all examples into tests. This means that if you change the underlying code now your test fails. And so this means that you can't even you can't forget to update your examples in your documentation.

    ↗Why Rust is different, with Alice Ryhlyoutube.com 4th of 23 in this recording

  45. AR

    Alice Ryhl quoted

    Our readingA design proposal should describe the feature as though it already existed — once as a tutorial and once as reference documentation.

    design

    Their wordsAnd what this is is that in the guide level explanation, you explain your feature as if you were writing a guide as if the feature already existed. And in the reference level explanation, you explain your feature again as if it already existed, but as if it would be in the language reference instead of a tutorial.

    ↗Why Rust is different, with Alice Ryhlyoutube.com 16th of 23 in this recording

  46. 3 weeks earlier
  47. PM

    Patrick McKenzie quoted

    Software engineer and writer, known online as patio11

    Our readingComplex professional practice, including law, can be reduced to written playbooks that make its outcomes reproducible without requiring artisanal mastery.

    Their wordsBut in law, as in restaurants, we are allowed to write down recipes and tell people to just follow the steps.

    ↗Notes on a non-profit indicted for bank fraudbitsaboutmoney.com 2nd of 3 in this piece

  49. PM

    Patrick McKenzie quoted

    Our readingThe mandatory financial-transaction reporting regime functions as an end-run around the Fourth Amendment.

    Their wordsOne critique is that this regime is functionally an end-run around the Fourth Amendment.

    ↗Notes on a non-profit indicted for bank fraudbitsaboutmoney.com 3rd of 3 in this piece

  50. 1 day earlier
  51. AD

    Anil Dash quoted

    Technologist and writer, a blogger since 1999

    Our readingAuthoritarian governments seek to destroy or falsify public records, including digital ones.

    Their wordsThe traditional authoritarian impulse to destroy or falsify the public record has not spared the digital realm under the current administration.

    ↗Why are the Artemis II photos on Flickr?anildash.com 3rd of 3 in this piece

  52. 5 weeks earlier
  53. MN

    Mark Nottingham quoted

    Web infrastructure engineer and standards editor

    Our readingGrounding a chatbot in a specific set of documents allows a genuinely new way of interacting with them.

    chat interfaces

    Their wordsgrounding a chatbot in a specific set of documents allows you to interact with them in a genuinely new way.

    ↗Using AI to Evaluate Internet Standards (Part Two)mnot.net

  54. 5 days earlier
  55. AK

    Andrej Karpathy quoted

    Founding member of OpenAI and former director of AI at Tesla

    Our readingStop writing documentation for people: a library should ship markdown for agents, which then explain it to whoever asks.

    Their wordsIt used to be that you have documentation for other people who are going to use your library, but like you shouldn't do that anymore. Like you should have instead of HTML documents for humans, you have markdown documents for agents.

    ↗Skill Issue: Andrej Karpathy on Code Agents, AutoResearch, and the Loopy Era of AIyoutube.com 19th of 20 in this recording

  56. 10 days earlier
  57. LO

    Luis Ouriach quoted

    Designer Advocate at Figma as of March 2026

    Our readingThe importance of the docs written for design systems, in whatever form, is going to be the foundation of everyone's workflow going forward, and guidelines are becoming paramount.

    design

    Their wordsthe importance of those docs, whatever form that they were in before, is going to be the foundation of everyone's workflow going forward. They might just be written slightly differently and for a different format, for a different output, but guidelines are becoming paramount for everything we need to do

    ↗Luis Ouriach - How are design systems changing?youtube.com 1st of 19 in this recording

  58. 4 weeks earlier
  59. PS

    Peter Steinberger quoted

    Founded PSPDFKit in 2011 and ran it for a decade

    Our readingDo not run a personal agent on a cheap or local model: weak models are gullible and easy to prompt-inject.

    Their wordsThat's why I warn in my security documentation, don't use cheap models. Don't use Haiku or a local model. Even though I, I very much love the idea that this thing could completely run local. If you use a, a very weak local model, they are very gullible. It's very easy to, to prompt inject them.

    ↗OpenClaw: The Viral AI Agent that Broke the Internet - Peter Steinberger | Lex Fridman Podcast #491youtube.com 6th of 30 in this recording