Related posts
Iván Olivares Rojas
x.com
Your Harness matters, you need to build it or follow something, but you can't leave it to chance. If you are starting with AI Engineering, be sure you learn what is a harness, this is a good video to do understand why it matters. My personal recommend is: try to understand 40% of what is, then try to build it, and continues learning from there.
continues harness
The the words it uses that this site has seen
least often elsewhere. Posts are matched on those words alone —
nothing here is a summary of this one.
Sources
All Writing Korrents GitHub x.com Newsletter Blog Recommends Bluesky Mastodon Site
Top people
Radley Balko
Sebastian Raschka
Gergely Orosz
Tomasz Tunguz
François Chollet
Federico Viticci
Dex Horthy
Lilian Weng
Vicki Boykis
Dan Cederholm
Eugene Yan
Chris Wellons
Showing Profile →
Show everything
Hiding Show them again
Show them again
Further back ↓
Hiding Show them again
20 September
Software engineer who writes about low-level and systems programming at nullprogram.com; author of the Elfeed feed reader, the Endlessh SSH tarpit and Enchive.
19 September
Writes The Pragmatic Engineer, the software-engineering newsletter, and wrote The Software Engineer's Guidebook. Formerly an engineering manager at Uber, and at Microsoft/Skype and Skyscanner before that.
Their words
having the very best AI model and harness doesn’t result in better/faster execution - in fact it slows you
Show the whole quote
@GergelyOrosz on X x.com
Political scientist writing Drezner's World on international relations, American politics and the marketplace of ideas; author of The Ideas Industry.
18 September
Indonesian software engineer, educator and engineering manager at Zero One Group; open-source enthusiast writing at ripandis.com.
17 September
Venture capitalist and founder of Theory Ventures, previously a partner at Redpoint. Writes a near-daily blog on startups, SaaS metrics, data and AI.
The better the harness, the better the business. Berkeley published a study this week showing harnesses, the systems that control AI agents, set the price of an answer. The right harness cuts the cost of the same result by 71% without a loss of accuracy. Related
Investigative journalist on policing, forensics and criminal justice; wrote Rise of the Warrior Cop and writes The Watch.
Venture capitalist and founder of Theory Ventures, previously a partner at Redpoint. Writes a near-daily blog on startups, SaaS metrics, data and AI.
The Harness Margin Opportunity A UC Berkeley study finds the harness sets the price of an answer : GPT-5.6 Sol costs 71% less on Pi than on Claude Code, the same model returning the same result, & none of the 42 harness comparisons shows a statistica…
Anthropic OpenAI Related
Engineer at Vercel Labs as of 2026. Ships a steady stream of small developer tools — agent-browser, portless, json-render, native-sdk, the v0 SDK — and lists every one of them at ctate.dev.
Their words
Every agent harness can now use Jev npm install -g ai-cli → ask yes/no questions → choose between options → score against your criteria https://t.co/dpgmcdCj8F
Show the whole quote
x.com
16 September
Austrian developer. Created the libGDX game framework, and more recently the pi coding agent, which moved with them to Earendil — Armin Ronacher's company — in April 2026. Writes at mariozechner.at.
imma regret saying this. but. pi 3.14. next week friday. just a new harness. Related
Writes The Diff, a newsletter on inflections in finance and tech read by hedge fund managers, founders and VCs. Previously worked in finance at SAC Capital/Point72.
PacePocalypse is over. SaaSpocalypse is back. The pick-your-'pocalypse market continues. Related
15 September
British comics writer and novelist (Transmetropolitan, Planetary, The Authority, Gun Machine). Keeps a public daily notebook logging what he reads, watches and listens to.
15sep26 New pens have arrived. Pigma Microns, Pentel Energels, a Uni-ball Signo 207, WRITECH Space-Up, and a new Drehgriffel gel pen. OPERATIONS: sprint week continues READING:
Related
13 September
Moroccan senior frontend developer writing at smakosh.com; builds side projects and client work under Smakosh LLC.
You could earn more by quitting your boring job at a big corp and start going after bug bounties with higher rewards using a good harness from the same company Related
Dutch bootstrapper who builds and runs their internet startups alone — Nomads.com, Remote OK, Hoodmaps, Photo AI and Interior AI — and has written about it at levels.io since 2013. Posts as @levelsio.
startups
Their words
In the most early adopting tech pioneering place that's Silicon Valley new startups aren't even building software anymore And it's debatable if anyone will actually need a custom harness or it won't just be generically offered by the AI frontier companies So software is mostly dead and hardware it is
Show the whole quote
levels.io
12 September
Former L8 engineer at Meta, Microsoft and Atlassian, now writing and building solo on agentic engineering. Writes Kun's Field Notes and posts a lot about AI coding agents on X.
every model provider should learn from meta's pricing model for muse spark, with fully transparent subsidization labeled for the contributor tier stop doing "in order to use our model at a good price, you must use our harness which secretly collects your data and/or upload your codebase" meta clearly proved there's a way to collect training data without relying on a harness. just be transparent about when you collect data vs not, clearly communicate how much the data is worth, and give the choice to the user and i really don't need your harness. i already have SO MANY state of the art harness… Related
Writes The Pragmatic Engineer, the software-engineering newsletter, and wrote The Software Engineer's Guidebook. Formerly an engineering manager at Uber, and at Microsoft/Skype and Skyscanner before that.
11 September
Economics professor at the University of California, Berkeley and a former Deputy Assistant Secretary of the US Treasury. Author of Slouching Towards Utopia and writer of the Grasping Reality newsletter.
Writes Stratechery, a subscription newsletter on the business and strategy of technology, and hosts the Dithering and Sharp Tech podcasts.
Co-founder and CEO of Stripe and co-founder of the Arc Institute. Keeps a personal site of reading lists, open questions and notes on scientific progress and how research is funded.
Their words
Chris Arnade's blog continues to be one of the very best: https://t.co/tUMFIUq8R7.
Show the whole quote
x.com
9 September
AI research engineer working on large language models. He writes the Ahead of AI newsletter and is the author of Build a Large Language Model (From Scratch).
8 September
Computer science professor who works on fast data processing; co-author of the simdjson parser and a weekly blogger about software performance since 2004.
PISA is a long-standing set of tests that are taken by students all over the world. It can be used to compare countries and to see how countries evolve over time. In Canada, we have results per province. In math, Quebec is usually best, and it is still the case for the latest results (2025). Yet the numbers fell. You see a clear drop starting in 2022, and it continues. The drop is worldwide. It is not wealth. Most countries got slightly richer, or had relatively modest decline. Quebec declined economically between 2022 and 2025 by about 1,2% (real dollar GDP). But it seems unlikely to be a go… Related
Machine-learning researcher on open language models; writes the Interconnects newsletter and the RLHF Book, after leading post-training at Ai2.
Latest open artifacts (#24): Motif-3, GLM-5.3, Hy4-preview and open model licenses The open model ecosystem continues to expand in its breadth The frontier labs continue to be enveloped in mass drama. Related
Mathematician and Senior Lecturer in the Mathematics department at Columbia University. Author of the book Not Even Wrong and of the long-running physics blog of the same name.
Creator of Ruby on Rails, CTO of 37signals, and creator of Omarchy.
Recommends their own rcmnd.app
Omarchy
Their words
Omarchy is going to be the best place to get every major AI harness and tool, configured with minimal effort
Show the whole quote
x.com
7 September
Founding executive editor of Wired and author of What Technology Wants and The Inevitable. Writes at kk.org and runs its long-running recommendation projects, True Films and Cool Tools.
6 September
Brazilian-American staff software engineer working on platforms, AI and developer productivity. Writes about engineering and his setup at ivansantos.me.
I had a great time building a @cursor_ai SDK provider for @opencode 2. If you got to use Cursor for free this past month, this is for you! Same harness I already live in, now with Cursor’s models under /models. Related
5 September
Writer and programmer; co-founder of the software studio Postlight and author of the Bloomberg Businessweek essay What Is Code?
Bluesky provides a mobile web view that runs great on my four-year-old phone and never pushes me to download an app. That is special and appreciated and I hope it continues. Related
3 September
Creator of the Keras deep-learning library and the ARC-AGI benchmark.
Their words
so harness capabilities are increasingly shifting into the model itself.
Show the whole quote
@fchollet on X x.com
Creator of the Keras deep-learning library and the ARC-AGI benchmark.
Their words
GPT-6 Astra represents a step-function change in model capability for interactive reasoning problems. It scores 66% on ARC-AGI-3 using our standard harness, and nearly 100% with a continuous conversation harness and custom compaction, at a cost of roughly $360 per game.
Show the whole quote
x.com
2 September
Technology writer of Spyglass, a newsletter about technology and media. Previously a reporter at TechCrunch and an investor at GV.
Product discovery coach; wrote Continuous Discovery Habits and writes Product Talk.
1 September
Writes Don't Worry About the Vase, a near-daily account of what is happening in AI and what they think it means. Former Magic: the Gathering pro and trader; unusually willing to put a number on a belief and to grade their own past calls.
AI alignment OpenAI
Their words
My worry continues to be that their fundamental approach is fatally flawed, and they are not focusing on the right things.
Show the whole quote
thezvi.substack.com
29 August
Developer and podcaster; created Instapaper and Overcast, co-founded Tumblr, and co-hosts the Accidental Tech Podcast.
I opened up Apple Maps' search box and its new ads suggested I go to JetBlue Airways, the JetBlue terminal at JFK airport 39 miles away. I opened up Google Maps' search box, and 100% of the screen space is devoted to my presets (Home/Work/etc.) and my Recent searches. Apple continues to embarrass itself with its shitty ad plays. In the meantime, I'll keep using Google Maps, because it's just *better*, and — something I never thought I'd say — it respects me more as the user, with fewer ads. 🤯 Google Related
Co-founder and CEO of Linear; previously principal designer at Airbnb, where he led the Design Language System, and founding designer at Coinbase.
Coding on @linear allows you to pick your harness + model Related
28 August
Canadian cartoonist and writer: Dinosaur Comics, Marvel's The Unbeatable Squirrel Girl, and the choose-your-own-path books To Be or Not To Be and Romeo and/or Juliet.
26 August
Full-stack developer, creator of Tailwind CSS; writes at adamwathan.me and makes courses and screencasts for web developers.
.@AmpCode is consistently 3-4 months ahead of every other harness in terms of discovering and optimizing the next workflow leap, feels like their only disadvantage is not being a model company and not being able to subsidize tokens. Insane to me that a model company hasn’t scooped them, would be a huge advantage now that most of the frontier models are equally capable. Related
22 August
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
20 August
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
19 August
Technical educator, conference speaker and developer relations engineer based in Ottawa, Canada. Author of the Anubis bot filter and of over 400 articles at xeiaso.net.
10 August
Founded PSPDFKit in 2011 and ran it for a decade. Came back from a break to work on AI agents — the OpenClaw project, and OpenAI, joined in February 2026. Writes at steipete.me.
27 July
Creator of Claude Code at Anthropic.
24 July
Co-founder of Modem and a founding engineer at Sentry, where he went on to be VP of Engineering. Co-author of Third-party JavaScript, and co-host of the State of Agentic Coding podcast with Armin Ronacher.
✅ Yes the code matters In experiments we ran, well-named and organized code can dramatically improve agent outcomes: - fewer tokens spent/more context window preserved - fewer confidently wrong answers developed Verified against 14 model/harness combos incl Sol, Fable, GLM 5.2 and more Quoting @benv.ca 👋 New blog post: "How coding agents read files – and how to write for them" In it I explain how code optimized for agent-discovery can lead to major token savings on search + retrieval, even with frontier models like GPT-5.6 Sol Plus other important findings ⬇️ Related
22 July
Security engineer; founder of Matasano Security and Latacora.
OpenAI
Their words
I genuinely believe that if you took an open weights model from 2025 and built a pentest harness for it, it could do this kind of sandbox escape and scan/hack in most networks. This is only surprising because you assume OpenAI has sounder sandboxes.
Show the whole quote
@tqbf on X x.com
21 July
Monetary economist known for NGDP-level targeting. Wrote TheMoneyIllusion until it closed in 2024 and now writes The Pursuit of Happiness.
15 July
Founder of HumanLayer; wrote 12-Factor Agents, on how to build AI agents that hold up in production.
From one piece
Context engineering with Dex Horthy
2 beliefs, in the piece's order there
14 July
Software design writer and chief scientist at Thoughtworks; wrote Refactoring and Patterns of Enterprise Application Architecture.
DSLs Enable Reliable Use of LLMs LLMs generate code incredibly fast, but to ensure they generate exactly what is intended, they need clear boundaries. Abstractions and Domain-Specific Languages (DSLs) provide a strong harness that guides LLMs right fro…
LLMs Related
Machine-learning researcher; has written the Lil'Log survey posts on how a model technique works since 2017, and worked at OpenAI from 2018 to 2024, latterly leading its safety systems team.
6 July
Writer and teacher on productivity and personal knowledge management. Author of Building a Second Brain and founder of Forte Labs, which runs the course of the same name.
4 July
Machine-learning researcher; has written the Lil'Log survey posts on how a model technique works since 2017, and worked at OpenAI from 2018 to 2024, latterly leading its safety systems team.
27 June
AI research engineer working on large language models. He writes the Ahead of AI newsletter and is the author of Build a Large Language Model (From Scratch).
25 June
Investigative journalist on policing, forensics and criminal justice; wrote Rise of the Warrior Cop and writes The Watch.
15 June
Machine learning engineer working on recommender systems, personalization and information retrieval. Previously worked on LLMs and LLM infrastructure at Mozilla.ai and on ML and recsys at Duo, Tumblr, Automattic and Comcast; wrote the 'What are embeddings?' text and ran the Normconf conference.
Their words
For my local setup, I'm currently using Pi as the agent harness and LM Studio as the inference server
Show the whole quote
vickiboykis.com
Their words
For my local setup, I'm currently using Pi as the agent harness and LM Studio as the inference server
Show the whole quote
vickiboykis.com
9 June
Web designer; founder of SimpleBits, co-founder of Dribbble, and author of Bulletproof Web Design and Handcrafted CSS.
Mastodon Work continues on Kyroh. Movies set in Egypt? I haven’t seen all of these, and I’m sure I’m missing some obvious ones. Haven’t kerned this one yet in case that’s not obvious! ;) Related
Bluesky Work continues on Kyroh. Movies set in Egypt? I haven’t seen all of these, and I’m sure I’m missing some obvious ones. Haven’t kerned this one yet in case that’s not obvious! ;) Related
5 June
Investigative journalist on policing, forensics and criminal justice; wrote Rise of the Warrior Cop and writes The Watch.
31 May
Statistician at Carnegie Mellon; writes on complex systems, machine learning and inference, and has kept a books-read notebook since 2000 that is now thousands of entries long.
Their words
Mind candy: continuing Robins's series of historical mysteries set in a very-slightly-alternate Regency London. (I re-read the last one, Sleeping Partner, before turning to this.) Robins continues to be great and I hope that many more of these will flow from her pen.
Show the whole quote
bactra.org
30 May
Author of Recoding America; founded Code for America and was US deputy chief technology officer from 2013 to 2014. Writes Eating Policy on why governments cannot do what they decide to do.
27 May
Builds developer tools — SST, and now OpenCode at Anomaly. Posts constantly on x.com and writes almost never; the blog stopped in 2021.
18 May
Member of technical staff at Anthropic. He has led ML/AI teams at Amazon, Alibaba and Lazada, and writes about LLMs, recommender systems and engineering at eugeneyan.com.
Cloudflare on their vulnerabilty discovery harness • Recon: Read the codebase, return an architecture doc • Hunt: ~50 agents look for bugs concurrently • Validate: Independent agents try to disprove findings • Gapfill: Areas that need a 2nd pass are flagged • Dedup: Findings with the same root cause combined • Trace: Confirms if attacker input reaches the bug • Feedback: If reachable, becomes new hunt tasks • Report: Write report with predefined schema blog.cloudflare.com
Related
28 April
Founded CD Baby in 1998 and sold it in 2008, giving the proceeds to a charitable trust for music education. Since then a writer and speaker: five books and 550+ articles, published first at sive.rs, which they hand-code and keep free of advertising.
I have so much more to say on this subject, but this book is done now because I believe short books are useful. So the conversation continues on the website: Related
22 April
Researcher on distributed systems and local-first software at the University of Cambridge; wrote Designing Data-Intensive Applications.
Their words
Scalability is just like what mechanisms we have for dealing with changes in load. If load increases, how can we add computing capacity to a system, for example, so that the system still continues working?
Show the whole quote
youtube.com
7 April
AI research engineer working on large language models. He writes the Ahead of AI newsletter and is the author of Build a Large Language Model (From Scratch).
31 March
Programmer and teacher; writes on software, learning and the web, and makes the Coding Blocks video course.
24 January
Machine learning engineer and consultant focused on RAG and retrieval systems. He writes about applied AI engineering at jxnl.co and is the author of the instructor library.
Their words
I've worn my Wyatt harness boots for years and they're still pristine.
Show the whole quote
jxnl.co
7 January
Executive coach for product leaders. Previously a partner at GV (Google Ventures) and a product manager at Google. Writes at Bring the Donuts, including a standing essential-reading list for product managers.
Their words
can lead to imposter phenomenon. In this book, a leading psychologist looks at the science behind our inner voice and new research into how to harness it and improve your physical and mental health.
Show the whole quote
bringthedonuts.com
1 December 2025
Investment researcher and engineer who runs Lyn Alden Investment Strategy, writing on macroeconomics, currency systems and equity analysis.
Why Most Cryptocurrencies Won’t Accrue Value December 2025 For many years now, I’ve been bullish on bitcoin, and continue to be, but have not extended that view toward any other cryptocurrency. That decision continues to hold up well. Since I first recommended it…
Related
23 November 2025
Member of technical staff at Anthropic. He has led ML/AI teams at Amazon, Alibaba and Lazada, and writes about LLMs, recommender systems and engineering at eugeneyan.com.
16 May 2024
Software engineer and writer; former CTO of Wave, now at Anthropic. Blogs at benkuhn.net on attention, engineering management and doing hard things well.
Talent density at Anthropic continues to increase :) Quoting @mikeyk I'm thrilled to announce that I've joined @AnthropicAI as their Chief Product Officer! The team at Anthropic is exceptional, and I felt at home from my first conversations with them. Daniela, Dario, Jared and the team embody the combination of deep talent, high empathy, and pragmatism that I loved… Anthropic Related
9 May 2023
Designer, product manager, YouTuber and developer who has written about technology at Birchtree since 2010. He makes the Quick Reviews and Quick Notes apps and the A Better Computer YouTube channel.
Their words
Professional video editing has never been more accessible than it is today, and Final Cut Pro continues to be my app of choice here.
Show the whole quote
thesweetsetup.com
28 August 2022
American software engineer working across web, mobile and TV platforms; writes at sean.is.
Just when I thought it was impossible to accumulate even more hate for them, Philadelphia Parking Authority continues to impress. Went to dispute a ticket and noticed the PPA website is scrolljacking 🙄 Related
Nothing matches. Show everything
A korrent is a belief a person has stated in their own words: one
sentence stating the claim, backed by a quote and a source, kept at
korrents.com .
Under a name here, the quoted block is what they actually said.
The korrent beneath it is the claim those words support, in
korrents' wording — tap it to see the record, its source, and who
else holds it.
Nobody here wrote their own korrents. They are compiled from public
statements, and a person can change their mind, which is recorded too.
Got it
About the English under a post
Some people here publish in a language other than English. Where they
do, this site shows a machine translation beneath the post, in
this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the
serif above is exactly what the person published, and it is what to
quote them on. A translation can be wrong in ways that matter,
especially about tone.
Only the post's own words are translated. A quoted post, a linked
article and a belief on korrents.com
are left in their original language.
Got it