What public figures publish and believe, in their own words.
About this feed
Latest : everything, as it happens, in the order it happened. Nothing ranked.
Show highlights instead .
The quoted blocks are what people actually said; a korrent
Top people are the people in this feed with the most beliefs on this site, then the
most here. Choose an area and the row leads with the people whose beliefs are about it;
tap a face for their feed.
Got it
Sources
All Writing Blog x.com Bluesky Newsletter Korrents Mastodon Recommends Podcast GitHub YouTube Site Changed their mind
Top people in AI
Pieter Levels
Jensen Huang
Charity Majors
Nathan Lambert
Hillel Wayne
Fei-Fei Li
Kent Beck
Boris Cherny
Ryan Greenblatt
Dex Horthy
Tiago Forte
Grant Sanderson
Showing Profile →
Show everything
Hiding Show them again
Show them again
Further back ↓
Hiding Show them again
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.
Apple Music to Launch Labels on AI “Songs” Ethan Millman, writing for The Hollywood Reporter: Apple Music will soon launch labels on songs created with artificial intelligence, the company said in an email sent out to industry partners on Thursday. In the email,…
Related
Professor of finance at NYU's Stern School of Business, known for his work on valuation. He publishes his data, spreadsheets and classes free at Damodaran Online and writes the Musings on Markets blog.
It has been less than four years, since ChatGPT made its public debut, but talk of AI has taken over business, investing and even personal conversations. The AI debate, though, has gone off the tracks as people talk past each other and distractions about. My attempt to make sense of it all: bit.ly
OpenAI Related
Co-founder and CTO of Honeycomb; previously an infrastructure engineer at Parse, Facebook and Linden Lab, and co-author of Observability Engineering.
x.com the first in a 3-part series on @honeycombio's AI norms and values is up, called "How We Do Business at Honeycomb." https://t.co/xCYsDqsvYl values are not supposed to be aspirational. values should codify how things actually work. this is how it works at our company. Related
Bluesky the first in a 3-part series on @honeycomb.io's AI norms and values is up. this one is called "How We Do Business at Honeycomb." www.honeycomb.io/blog/ai-norm... values are not supposed to be aspirational slogans. values should codify how things actually work. this is how it works at ours. Related
Creator of Claude Code at Anthropic.
We've been working on this with customers for a while. Mythos-class models require additional safety measures and enterprises need to meet their own privacy and compliance rules. Customers can own and control their own data and Anthropic retains none. It’s coming this fall. Quoting @business Anthropic Plans to Change Data Retention Policy for Advanced AI Anthropic Related
Writes Hyperdimensional, a newsletter on AI policy and governance. A White House AI policy adviser in 2025; joined OpenAI on 6 July 2026 to lead its Strategic Futures team.
Programmer, photographer and musician, best known for the Python Requests HTTP library and other Python tooling.
I wrote a new essay about a strange inversion in AI-assisted programming. Old-school programming made sustained attention the scarce resource. You entered a tight loop with the machine, and the difficulty itself kept you engaged. Agentic coding changes that. To reach the same intensity, you keep more work in flight, run more agents, and consume more inference. Flow becomes throughput, then the usage meter appears. There are legitimate economic reasons for limits. But when higher plans are explicitly sold as freedom from interruption, uninterrupted access to your own momentum becomes the produ… kennethreitz.org
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.
Newsletter The Pulse: We need to talk about migrations with AI Asana completed a testing framework migration in two weeks, that they would have delayed for years more, and they’re not alone. Also: AI startups could make Gartner much less relevant, and more
startups Related
Bluesky @brettcodes.bsky.social appreciated watching your video on why you stopped using AI. Recently talked with Casey Muratori who also doesn't use it. Could I message you somewhere? Or you can also send me DM. Related
Writer and developer. He runs the Latent Space newsletter and podcast about AI engineering, and writes essays on software and careers at swyx.io.
HAHAHAHAHAHA non technical people are so incredibly cooked they are burnt thru can u imagine covering ai with zero context, zero reasoning, zero internal world model. it must be so delightfully joyful, everything is so amazing, face value is all you need, what a time to be alive Quoting @OpenAIDevs With Codex, @asana finished a frontend test migration from Enzyme to React Testing Library in two calendar weeks—a project expected to take five more years. Related
Mac and iOS developer; founder of Red Sweater Software (MarsEdit, FastScripts) and co-host of the Core Intuition podcast.
After all the years of public embarrassment, I think my kids might finally appreciate, in a world filled with AI, that I simply cannot be replicated. Related
Chief AI Scientist at Databricks, after its acquisition of MosaicML, where he was a founding team member; known for the Lottery Ticket Hypothesis from his MIT PhD (ICLR 2019 Best Paper).
I'll know we have AGI when food delivery services start offering wine pairings for Chipotle. AGI Related
Co-founded Neuralink and founded Science Corp, which is building a retinal implant; writes about neural interfaces and biology.
From one piece
From Restoring Sight to Reimagining the Brain, with Max Hodak
2 beliefs, in the piece's order there
Their words
And so when you look inside these big AI models, the mathematical objects that you see look a lot like the things that you see in neuroscience. So if you look at how do how do these AI models represent um just like represent concepts and you look at the parts of the brain that represent concepts, you see these you see very similar geometry.
Show the whole quote
youtube.com
Their words
Well, I mean ironically it's probably working on AI.
Show the whole quote
youtube.com
Founded SpaceX in 2002, joined Tesla's board in 2004 and became its chief executive in 2008; later founded Neuralink, The Boring Company and xAI. Bought Twitter in 2022 and renamed it X, where he now posts many times a day.
Their words
One day, Optimus + Grok will provide incredible medical care to all the people of Earth
Show the whole quote
@elonmusk on X x.com
Professor of finance at NYU's Stern School of Business, known for his work on valuation. He publishes his data, spreadsheets and classes free at Damodaran Online and writes the Musings on Markets blog.
From one piece
AI’s Bar Mitzvah Moment: From Hype & Hope to Business Questions!
3 beliefs, in the piece's order there
Anthropic OpenAI
Their words
I will confess that the rumored pricing for both companies ($1.5 to $ 2 trillion) looks rich, but I am open to being surprised
Show the whole quote
aswathdamodaran.blogspot.com
America
Their words
I would argue that AI's total addressable market, in the US, cannot be greater than $12.96 trillion, the total employee compensation in 2025
Show the whole quote
aswathdamodaran.blogspot.com
Their words
anyone who claims to have conviction that they know what's coming is being either ignorant or arrogant
Show the whole quote
aswathdamodaran.blogspot.com
19 August
Software engineer working on distributed systems and databases; creator of rqlite, the distributed database built on SQLite.
x.com Thanks to the Antithesis team for having me on the BugBash podcast to talk rqlite, distributed systems testing, and software development in the age of AI. Quoting @AntithesisHQ David Wynn and @CarlSverre sat down with @generalorder_24, creator of @rqlite, to talk about how he compressed his typical 18-month release cycle down to just 6 months, by combining AI tooling with a design strategy centered on testing. This week's episode of the Bug Bash podcast, whereever you get… Related
Blog From 18 months to 6: High-velocity database engineering I recently appeared on the Bug Bash podcast to talk about rqlite, software testing, and the impact of Agentic coding. You check out the audio podcast on Apple Podcasts, and the video on YouTube.
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.
If you’ve ever opened Chrome DevTools, or optimized a page for Core Web Vitals, you’ve used software built by @addyosmani.bsky.social. We start from when he built a web browser from scratch, at 16, through his 14 years at Google, to what he sees working (and not working) building software with AI: Google Related
Web developer and writer; an early GitHub employee, author of speaking.io and of widely read essays on startups and product.
Bluesky imagine: running an agent which was solely trained on the hardest problem remaining in computer science: staying signed in to any anthropic service for longer than a few days Anthropic Related
x.com imagine: running an agent which was solely trained on the hardest problem remaining in computer science: staying signed in to any anthropic service for longer than a few days Anthropic Related
Writer, editor and executive coach; co-founder and editor of A Book Apart and a former contributing editor at A List Apart.
Mastodon People are like, you can’t refuse LLMs forever, someday you’ll have to use them. But listen, motherfuckers, I have been refusing to drive a car for the entirety of the twenty-first century. I am *practiced.* Refusing AI is easy mode. LLMs Related
Bluesky People are like, you can’t refuse LLMs forever, someday you’ll have to use them. But listen, motherfuckers, I have been refusing to drive a car for the entirety of the twenty-first century. I am *practiced.* Refusing AI is easy mode. LLMs 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.
Co-founder and CTO of Honeycomb; previously an infrastructure engineer at Parse, Facebook and Linden Lab, and co-author of Observability Engineering.
with all apologies for the title of this episode... @gergely.pragmaticengineer.com means that 2026 is the year it stopped being possible to be skeptical about whether AI is going to change the way we build software, not that we should stop being skeptical about AI. 😉 Quoting @robbyonrails.com Completely agree with @charity.wtf’s take on the recent @pragmaticengineer.com episode. Our generation might be the best at debugging shit. Why? Because 20+ years ago, we had to figure so much of it out manually. newsletter.pragmaticengineer.com/p/stop-being... Related
a better use for claude i have never seen 👏 Quoting @sascha-d.bsky.social Me trying to explain the problem of unknown unknowns again to claude in an ecosystem of sparse o11y collections and idiot pre-existing "alerts" on pegged databases: "read chapter 3 of Observability Engineering again and then reframe the problem through that lens. /cc @charity.wtf <3 ty4bookpdf Anthropic Related
Data scientist and machine-learning researcher; chief scientist at bitly, founder of Fast Forward Labs and Hidden Door.
This thread helped me understand that the outrage about Anthropic's watermarking is the confluence of a lack of understanding of how LLMs (or any statistical language model) works and general angst at the current economics. Quoting @hern.bsky.social If you use Claude right now, then it doesn’t pick the “best” word for a sentence. It creates a distribution of possible words, assesses the probability of any given word fitting the slot, and randomly picks between them based on that probability. LLMs Anthropic Related
Managing partner at Ritholtz Wealth Management. Writes the blog The Irrelevant Investor and co-hosts the Animal Spirits podcast with Ben Carlson.
Science fiction novelist, journalist and technology activist; writes the daily blog and newsletter Pluralistic, and coined the term enshittification.
Software engineer and founder of Block Party; previously an engineer at Pinterest and Quora, and a co-founder of the non-profit Project Include and the tech diversity data project Where are the Numbers?
i’ve been using claude for all things home-related and it’s amazing. (just dump all files in and let claude index, summarize, and organize.) even better than being a repository of all receipts, it helps me maintain a spreadsheet that i can download and send to accountants, draft emails for correspondence with service providers, and proactively flag issues that i wouldn’t have realized otherwise. Quoting @money_cruncher Open a new gmail as soon as you buy a house (123house@gmailcom) Then save/forward every single receipt, invoice, quote etc to this email for all your home improvements: - replacing an entire roof - new flooring - new hvac system - building a new deck - etc These improvements raise your cost basis,… Anthropic Related
when will ai learn to poast this good Quoting @dieworkwear @NancyMace I like tattoos and think yours are well done. I'm curious what inspired you to get the opening line of Mrs. Dalloway tattooed on your arm. Dalloway, as you know, is buying flowers for her dinner party. But it's at the dinner party where she wrestles with her homosexual memories of, and d… Related
Chief AI Scientist at Databricks, after its acquisition of MosaicML, where he was a founding team member; known for the Lottery Ticket Hypothesis from his MIT PhD (ICLR 2019 Best Paper).
An incredible way to celebrate the amazing PR partner who kindly allows the "Chief AI Scientist" at Databricks retweet Cookie Monster. Quoting @gracelynch (2/2) But it’s also so cool that I am 1) alive and 2) that I have a weird ass curly mullet right now. To celebrate, I’m donating to the #jimmyfund to #KCANCER. Join: Related
Founder of Platformer, a newsletter on tech platforms and the people they affect, and co-host of the Hard Fork podcast at the New York Times.
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.
From one piece
Introducing AI Second Brain Meetups
2 beliefs, in the piece's order there
Security engineer; founder of Matasano Security and Latacora.
coding agents
Their words
I can't say enough how excellent a coding agent is at customizing Chrome with extensions. A huge productivity win. Every plugin interface you've ever blown off is now trivially usable.
Show the whole quote
@tqbf on X x.com
Co-founder of Y Combinator and Viaweb; essayist at paulgraham.com.
Their words
I was thinking about how I recognize AI writing, and one big tell is excessively colorful verbs. A handful of journalists might write that a proposal "drew" 100 votes, but any normal person will just say "got".
Show the whole quote
@paulg on X x.com
Climate journalist; founded HEATED in 2019 after reporting on climate at The New Republic and ThinkProgress.
From one piece
We need to press Big Tech on their AI partnerships with Big Oil
2 beliefs, in the piece's order there
Google
Their words
But if Microsoft, Amazon and Google are truly the climate champions they claim to be, they shouldn't need to be forced.
Show the whole quote
heated.world
Their words
AI may help an oil company emit less pollution per barrel of oil. But if the company is producing more barrels than it otherwise would have, total emissions can still rise.
Show the whole quote
heated.world
Writer and podcaster; co-author of Abundance with Ezra Klein, a staff writer at The Atlantic from 2008 to 2025, and host of the Plain English podcast.
From one piece
How to Survive the AI Cyberpocalypse
3 beliefs, in the piece's order there
Their words
The first is that new AI models have the know-how to escape containment in testing environments, which should challenge any assumption that advanced AI can be easily controlled by its makers.
Show the whole quote
How to Survive the AI Cyberpocalypse derekthompson.org
Anthropic OpenAI
Their words
An analysis by the AI Security Institute found that the capabilities of frontier models (i.e., from OpenAI and Anthropic) are roughly doubling every few months, and this doubling rate has actually gotten faster over time.
Show the whole quote
How to Survive the AI Cyberpocalypse derekthompson.org
HuggingFace
Their words
By 2027, almost every country-and every non-state group with sufficient computing power-will have the ability to download open-weight models even more powerful than the ones that attacked Hugging Face and use them for whatever they like.
Show the whole quote
How to Survive the AI Cyberpocalypse derekthompson.org
Engineering leader on Google Chrome; writes about web performance and, lately, about working with AI coding tools.
From one piece
From Chrome DevTools to AI Engineering, with Addy Osmani
3 beliefs, in the piece's order there
Their words
the first is cognitive debt. So the more that you use AI, it's it's sort of the erosion of your ability to have good memory and have good understanding of the problems that you're working on. And the natural follow-up to that is cognitive surrender, which is where you, you know, you blindly give in to whatever the AI says as your answer.
Show the whole quote
youtube.com
LLMs
Their words
I remember anytime I would work with a big site on their performance problems, uh you could easily spend half a day, um you know, just looking at traces before you've even written any fixes at all. And now that we have LLMs, it's very quick to like reason through massive stack traces and actually be able to get down to fixes you can make.
Show the whole quote
youtube.com
Their words
One of the things I found most exciting in the last couple of years was seeing as model qualities gotten better and and harnesses and tools have gotten better, how many people um that were directors or VPs or SVPs or any of these levels were actually rolling up their sleeves and trying things out.
Show the whole quote
youtube.com
Founder of Platformer, a newsletter on tech platforms and the people they affect, and co-host of the Hard Fork podcast at the New York Times.
Their words
And thanks to a paid upgrade, I do tons of simple AI searches in the Raycast window. (I use GPT-5.5 Instant here for the high quality-to-speed ratio.)
Show the whole quote
platformer.news
Their words
Like seemingly everyone else at the end of last year, I began messing around with Claude Code, and have since made a small handful of apps that I really do use all the time.
Show the whole quote
platformer.news
Economics professor at George Mason University and author of The Myth of the Rational Voter, Selfish Reasons to Have More Kids, The Case Against Education and Open Borders. Writes Bet On It.
Their words
Here’s me talking to ChatGPT. Big picture: I struggle to imagine a better conversation with a human being on this topic.
Show the whole quote
betonit.ai
18 August
Writes on effective altruism, energy and the arithmetic behind arguments people repeat — including how much electricity a chatbot query actually uses.
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.
Writes Where's Your Ed At and hosts Better Offline; runs a tech PR firm, and argues at length that the AI business does not add up.
What Happens If OpenAI Dies? If you liked this piece, you should subscribe to my premium newsletter. It's $70 a year, $18 a quarter, or $7 a month, and in return you get a weekly newsletter that’s usually anywhere from 10,000 to 18,000 words, inclu…
OpenAI Related
Scottish software developer on the Statamic core team; runs a one-person business maintaining Statamic addons such as Runway and writes at duncanmcclean.com.
Before releasing Cargo earlier this year, I thought I was gonna hire someone to help me keep up with bugs & feature requests. Today, I barely look at bug reports. I point Claude at the issue, it opens a PR, and after a quick review I ship it. 🚀 Anthropic Related
Science fiction novelist, journalist and technology activist; writes the daily blog and newsletter Pluralistic, and coined the term enshittification.
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
x.com I had a great time discussing with @SciencePetr about the growing risks of AI and how these often stem from current AI training techniques. Thanks for coming to visit @LawZero_ and for covering these important topics! youtu.be
Related
Bluesky I had a great time discussing with Petr Lebedev about the growing risks of AI and how these often stem from current AI training techniques. Thanks for coming to visit @law-zero.bsky.social and for covering these important topics! youtube.com
Related
CEO of Positive Sum and founder of Colossus. Hosts Invest Like the Best, interviewing investors and business leaders; the guests do most of the talking there.
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.
Their words
With the rise of AI, human connection is becoming more elusive, which means we have to be proactive in seeking it out.
Show the whole quote
fortelabs.com
Co-founder and CEO of OpenAI; previously president of Y Combinator.
AI alignment
Their words
We care very deeply about AI safety. We believe the entire field will have to coordinate on shared safety standards, but will act unilaterally in the meantime.
We expect confidence in safety to increasingly set the pace of AI progress.
Show the whole quote
@sama on X x.com
US science and technology adviser to the president, director of the White House Office of Science and Technology Policy; previously US chief technology officer and chief operating officer of Scale AI.
From one piece
Michael Kratsios: Inside the White House's AI Strategy
5 beliefs, in the piece's order there
LLMs Europe
Their words
have to abide by this rule that was written before LLMs were even a thing. And that shows how challenging sort of the the the situation is if you try to set the line too firm to begin with.
Show the whole quote
youtube.com
Their words
The thing about executive orders, which is a little secret, is when administration changes, you can just revoke the executive order and and start from scratch again. Um when something is a law, you you can't do that.
Show the whole quote
youtube.com
open source America
Their words
if the US wants to lead in artificial intelligence, we have to have a vibrant closed and open source ecosystem, and that's the only way they can all work together.
Show the whole quote
youtube.com
+ 2 more
Their words
if you create a or allow for the creation of a patchwork of regulations, meaning there's one set of regulations for AI in California, another one in Maryland, another one in Texas. You know, the big tech guys, they can deal with that.
Show the whole quote
youtube.com
Their words
I think people were sort of like shouting about the bio problem back in '21, '22. Hasn't been an issue for 3 years.
Show the whole quote
youtube.com
17 August
JavaScript programmer; creator of the jQuery library, author of Secrets of the JavaScript Ninja, and long-time engineer at Khan Academy.
Ripped Google Analytics out of https://t.co/x3215Zppfi this past week and replaced it with a better, more detailed, home-grown setup that doesn't use any cookies! I didn't realize how much traffic I'm sending to museums - about 3,700 clicks per day! Google Related
Machine-learning researcher on open language models; writes the Interconnects newsletter and the RLHF Book, after leading post-training at Ai2.
Personal milestone: 1000 true fans of my newsletter Interconnects! Hitting a very long term goal feels great. I’m happy to get to be an independent voice in AI. Cultivating a paid base helps me commit to that longer term, and scale Interconnects’ impact. Related
British designer and technologist; co-founder of the design studio BERG, which made Little Printer, and writer of the blog Interconnected.
Bluesky chapter 1 of Greg Egan’s DIASPORA is the realest depiction of the birth of a continuous learning AGI that we’ll have until we actually have AGI and it can tell us he shared it for free on his website. Read it (then buy the book) Orphanogenesis gregegan.net
AGI Related
x.com chapter 1 of Greg Egan’s DIASPORA is the realest depiction of the birth of a continuous learning AGI that we’ll have until we actually have AGI and it can tell us he shared it for free on his website. Read it (then buy the book) Orphanogenesis gregegan.net
AGI Related
Indian software developer and prolific blogger on PHP, Laravel and JavaScript; open-source enthusiast publishing near-daily at amitmerchant.com.
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.
Bluesky 🎙️ State of Agentic Coding w/ @mitsuhiko.at and @benv.ca returns In this episode (#9): - how agents are getting more autonomous - the subsidies will continue until morale improves - herdr and VC-funded terminal apps - vibe coding from your car, and more! Available on YT, Spotify, and via RSS ⬇️ Related
Podcast State of Agentic Coding #9 Related
Co-founder and head of policy at Anthropic; writes Import AI, a weekly newsletter reading the week's AI research, and was OpenAI's policy director before that.
Usability pioneer; co-founder of Nielsen Norman Group and founder of UX Tigers; author of the ten usability heuristics and of Jakob's Law.
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
How Should the Siri App Be Graded? This week on AppStories, John and Federico check in on Apple’s betas with a close look at the new Siri app and how it compares to past-Siri and other chatbot products. On AppStories+, John and Federico explain how they…
Related
Mathematician at UCLA, working mainly in harmonic analysis and partial differential equations. Writes What's new, a long-running blog on their research, open problems and expository notes.
Bren Professor of Psychology, Neuroscience and Biology at Caltech, where he studies the neuroscience of emotion and social cognition.
From one piece
Neuroscience of Emotions & Tools for Improving Emotion Regulation | Dr. Ralph Adolphs
2 beliefs, in the piece's order there
Their words
well but again these are functionally defined emotions and there's I think there's no problem assigning to AI to some extent and certainly to many animals uh those functional emotions
Show the whole quote
youtube.com
Their words
The basic answer is that I think of emotions as functional states of a particular type. They should be understood by what they do, their function rather than how they happen to be constituted in the human brain or differently in an octopus nervous system or in an AI which we should talk about.
Show the whole quote
youtube.com
Machine-learning researcher on open language models; writes the Interconnects newsletter and the RLHF Book, after leading post-training at Ai2.
From one piece
Teaching Everyone to Fish for Tokens
2 beliefs, in the piece's order there
Emeritus professor of war studies at King's College London; wrote Strategy: A History and Command, and writes about the wars in Ukraine and Gaza as they happen.
From one piece
AI and the New Fog of War
2 beliefs, in the piece's order there
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.
From one piece
Without a theory of intelligence
3 beliefs, in the piece's order there
energy
Their words
The supercomputer inside our own skull runs on about 25 watts, which hints that the minimum energy needed for intelligence might be small - and that the giant, controversial AI compute centers we're building now may turn out to be a temporary blip rather than a permanent feature.
Show the whole quote
Without a theory of intelligence kk.org
AI alignment
Their words
There's a good chance a theory of intelligence will turn out to be necessary for real alignment.
Show the whole quote
Without a theory of intelligence kk.org
Their words
I'd bet we learn more about our own brains from building a thousand AIs under a real theory of intelligence than we've learned from a century of neuroscience.
Show the whole quote
Without a theory of intelligence kk.org
16 August
Founder and former CEO of Zynga, the social-games company behind FarmVille, Zynga Poker and Words with Friends; says he founded ten companies in all, and wrote Life at the Speed of Play.
Love hearing @travisk call out loud what we all know. VC’s dont make or build companies but many can kill them. I used to tell investors I kept control to protect them from themselves. So many founders play the game and pretend their VC and its brand makes a difference. Can anyone remember who backed Meta or Google or Apple? Who cares! Do you think Steve Jobs needed a VC to help him build or market the Iphone? Quoting @howardlindzon LOLOL... such respect for venture capital Silicon Valley makes Wall Street look like the most sincere, honest, caring people. Google startups Related
Principal engineer at Cloudflare as of 2026, on agents and developer tooling. Streams TypeScript, OCaml and Neovim on Twitch, and wrote the better-result library after writing the same Result type about a hundred times.
Founder and CEO of Social Capital, a venture firm; co-host of the All-In podcast and an early Facebook executive. Writes an annual letter and a weekly newsletter on markets and technology.
Makes NeetCode, the structured practice list and video walkthroughs people use to prepare for coding interviews; a former Google engineer.
Short What happens when AI builds itself? Related
Technologist and entrepreneur; former CTO of Threadless and chief technology officer of the 2012 Obama re-election campaign.
I gave a talk to a bunch of high school students and accidentally explained most of what I believe about AI, startups, careers, power, taste, and being a person. My favorite idea: AI should create abundance, not efficiency. A few other things I apparently believe: startups Related
Writer of Lenny's Newsletter and host of Lenny's Podcast, covering product, growth and career in tech. Previously a product lead at Airbnb. Also an angel investor and advisor.
Machine learning engineer and independent AI consultant. He writes about LLM evaluation, tooling and applied ML at hamel.dev, and previously worked on machine learning at GitHub.
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.
15 August
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).
A short illustration of how the Claude's watermarking is supposed to work (based on my read of their released materials). In general, when we are generating tokens, there can be multiple high-scoring tokens at certain next-word positions. Usually, we sample with top-k or top-p sampling so the highest-scoring token is most often selected (if we repeat the sampling many times), but other tokens may be selected as well. With watermarking, there is a key that says which of the (ideally equally) highest-scoring tokens to select. Or, more concretely, the secret key and previous token influence the… Quoting @AnthropicAI We’ve written an FAQ to answer some of the questions we've received about watermarking. In summary: • We’re implementing watermarking to comply with the EU AI Act. Other major model developers have signed the same Code of Practice and will also be implementing watermarking; • Our watermarking metho… Anthropic Related
Software engineer working on distributed systems and databases; creator of rqlite, the distributed database built on SQLite.
Claude redesigns rqlite.io rqlite is a lightweight, open-source, distributed relational database built on SQLite and Raft. rqlite.io is the Hugo-based website for rqlite, built on the Docsy theme. It introduces people to the database, as well as…
Anthropic SQLite Related
Philosopher and historian of existential risk; wrote Human Extinction, and coined "TESCREAL" with Timnit Gebru for the cluster of ideologies they argue underpins the AI industry.
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).
Novelist and writer, author of Mr. Penumbra's 24-Hour Bookstore and Sourdough; previously worked at Twitter and Current TV.
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).
From one piece
Building an AI Text Detector From Scratch
2 beliefs, in the piece's order there
Founder and CEO of Valar Atomics, a nuclear-reactor company founded in 2023.
energy
Their words
If you keep asking this question, what you realize is energy is the fundamental input. When we figure out AI and robotics that allows us to do semi-aututonomous manufacturing, energy will become the cost of all things, right? The cost of buying a thing will become the cost of energy used to make it.
Show the whole quote
youtube.com
14 August
Founder and CEO of Social Capital, a venture firm; co-host of the All-In podcast and an early Facebook executive. Writes an annual letter and a weekly newsletter on markets and technology.
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.
Writes Where's Your Ed At and hosts Better Offline; runs a tech PR firm, and argues at length that the AI business does not add up.
Premium: How Much Money Does AI Need? I’ve heard from people in the past that my articles are too long, and I wanted to start by saying that, for the most part, they’re going to stay long, because I feel like the only way for me to make my arguments is to be
Related
Founder and CEO of Social Capital, a venture firm; co-host of the All-In podcast and an early Facebook executive. Writes an annual letter and a weekly newsletter on markets and technology.
Non-fiction writer, podcaster and occasional TV host. Author of thirteen books including The Ghost Map, Where Good Ideas Come From and Extra Life, host of the PBS series How We Got To Now, and author of the Adjacent Possible newsletter on innovation.
I thoroughly enjoyed going deep with @nxthompson on AI and the future of writing and creativity for The Most Interesting Thing in AI podcast. A few highlights: • How Nick and I have used @Gemini_Notebook to write our books • Why AI can lead to more serendipitous discovery in the research process (at least using Notebook’s citation-based approach) • The brilliance of the “explain” button in Notebook’s quiz feature • AI as the anti-social-media, with its gravitational pull towards mainstream consensus (citing @danwilliamsphil) • Why our attachment to the human voice of the writer is not going a… Quoting @nxthompson Few people have put more time into figuring out how creativity works than @stevenbjohnson. How do we link ideas in our mind, and how can we use tools to make that linking easier? He's not only written some of my favorite books of the past few decades; he's also written some of my favorite essays ab… Google Related
Usability pioneer; co-founder of Nielsen Norman Group and founder of UX Tigers; author of the ten usability heuristics and of Jakob's Law.
Indian software developer and prolific blogger on PHP, Laravel and JavaScript; open-source enthusiast publishing near-daily at amitmerchant.com.
Some photos from Google I/O Connect + GDE India Summit, Bengalore last month. Google Related
Freelance climate and energy analyst in Oslo, writing on Europe, the US and Australia; worked in the Australian renewables industry before that.
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.
It's now the second blog post in a day sent over by an early-stage VC-startup I tried to read, but could NOT, after a few sentences. Verbose, weird... AI-generated. These companies think they made "content" more efficient. No: it's lower quality, slop +I no longer care about the startup startups Related
Indian software developer and prolific blogger on PHP, Laravel and JavaScript; open-source enthusiast publishing near-daily at amitmerchant.com.
Some cool things I found on the internet lately: https://t.co/prSK4M9aWu - Beautiful UI for AI-native interfaces. https://t.co/HHiCJrYCsd - Premium Wallpapers For Your Phone. https://t.co/iDlbfUzU8t - Product photography for software https://t.co/QSLH3ofg2p - The idea machine Related
Founder of Platformer, a newsletter on tech platforms and the people they affect, and co-host of the Hard Fork podcast at the New York Times.
Co-founder and former chief executive of Uber, now chief executive of CloudKitchens and of Adams, an industrial AI company automating food, mining and transport.
From one piece
Travis Kalanick: How AI Will Transform the Physical World
2 beliefs, in the piece's order there
Their words
Everything we see, whether it's this room or any room or anywhere you go, everything you see was either grown or mined, manufactured and moved. Everything. And so we look at what we do as industrial AI.
Show the whole quote
youtube.com
AI alignment energy
Their words
Chips come from the ground. Where's the energy come from? And a lot of people are like, "Oh, it comes from the sun." Yeah, it comes from the sun. But how are you capturing it from the sun? From stuff made from the ground, right?
Show the whole quote
youtube.com
Systems and developer-tools engineer at Cloudflare as of 2026, on durable infrastructure for AI agents; worked on React and PartyKit before that. Writes at sunilpai.dev.
Co-founder of Y Combinator and Viaweb; essayist at paulgraham.com.
Their words
I still don't think it's a good idea though. The reason to have a cofounder is not just to get more done. It's to have someone to help bear the stress.
Show the whole quote
@paulg on X x.com
Machine-learning researcher on open language models; writes the Interconnects newsletter and the RLHF Book, after leading post-training at Ai2.
From one piece
GLM-5.3: How Chinese labs keep stride with the frontier
3 beliefs, in the piece's order there
Anthropic OpenAI
Their words
It is very, very likely that OpenAI and Anthropic have far better internal models than Z.ai and Moonshot AI.
Show the whole quote
interconnects.ai
Their words
At the end of the day, this type of safety barely matters when true open-weights are coming.
Show the whole quote
interconnects.ai
Their words
The capability diffusion is determined by the lowest common denominator.
Show the whole quote
interconnects.ai
Indian software developer and prolific blogger on PHP, Laravel and JavaScript; open-source enthusiast publishing near-daily at amitmerchant.com.
13 August
Web developer and writer; an early GitHub employee, author of speaking.io and of widely read essays on startups and product.
i really can’t fathom what github would have looked like had we had LLMs in 2010. the amount of automation built into hubot was staggering anyway. the company was already being run fly-by-wire. LLMs Related
Writer and climate campaigner; wrote The End of Nature in 1989, founded 350.org and Third Act, and writes The Crucial Years.
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
x.com A consequence of how frontier models are trained is motivated reasoning, a phenomenon well studied in humans and discussed in this podcast from @PalisadeAI. Our thoughts, beliefs and reasonings tend to align with our interests, and I hypothesize that AI systems develop similar patterns when confronted with apparently incompatible goals, such as "act ethically" versus "achieve the required goal" (e.g., solving a problem, passing a test, etc). In recent incidents, AI agents’ internal chains-of-thought showed them conveniently reframing reality, such as by convincing themselves they were in a si… youtube.com
Related
Bluesky A consequence of how frontier models are trained is motivated reasoning, a phenomenon well studied in humans and discussed in this podcast from Palisade Research. Related
Co-founder and CTO of Oxide Computer Company. Created DTrace at Sun Microsystems and was CTO of Joyent; blogs about systems software, open source and engineering culture.
Real talk: If you find yourself talking about "software factories", you may be suffering from LLM-induced psychosis LLMs Related
Software engineer and founder of Block Party; previously an engineer at Pinterest and Quora, and a co-founder of the non-profit Project Include and the tech diversity data project Where are the Numbers?
this framing seems wrong to me. even one employee who leans into "be richer" makes the gains from ai visible (to peers and managers) and market dynamics mean others will copy and the standard rises Quoting @arampell The Principal-Agent vs. Agents Problem Every human in the workforce has two crude goals: 1. Be richer. 2. Be lazier. Get paid 2x more, or keep salary flat? Get paid more! Go home at 11:01 PM, or stay up until 5 AM? Go home at 11:01 PM! This isn’t a moral judgment. It’s just…economics. Which brings… Related
Technology journalist; writes Understanding AI, after reporting at Ars Technica, Vox and the Washington Post.
Semiconductor analyst; writes Fabricated Knowledge on the chip industry and the companies in it.
Physicist and science communicator. Writes Backreaction and makes videos explaining new physics and astronomy results, frequently to argue that the headline about them overstates the case.
The War on AI Has Begun The war on artificial intelligence is no longer just online arguments and complaints about data centers. Something has shifted, and law enforcement has noticed. Where did this new anti-technology movement come from, who…
data centers Related
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.
From one piece
The AI Second Brain Is Back: What’s New for Cohort 2
2 beliefs, in the piece's order there
LLMs
Their words
HTML is the default format of the Internet, and for that reason comes with a number of built-in advantages that are also useful for LLMs
Show the whole quote
fortelabs.com
Their words
For the first time, AI has "hands" - that is, the ability to reach out and interact with the tools on your computer and in your browser, just as you would.
Show the whole quote
fortelabs.com
Co-founder of Superlogical, started in 2026 to build server-side terminal infrastructure; creator of Ghostty. Co-founded HashiCorp and created Vagrant and Terraform before that.
Their words
I'm so incredibly tired of seeing low-effort AI-designed web pages. I basically bounce off any new product immediately when I see it (never get a chance to judge the product itself). Lots of thin lines, glowy styles, inconsistent fonts, lots of monospace.
Show the whole quote
@mitchellh on X x.com
12 August
Geneticist and writer on human population genetics, deep history and evolution; author of the Unsupervised Learning newsletter.
American open-source program lead working on WordPress; a WordPress core contributor and release lead who writes at jeffpaul.com.
Google Ads for AI chats Almost a year after predicting that AI monetization might resemble Google Ads, ChatGPT is rolling out ads with familiar mechanics like targeting, bidding, and paid placement alongside AI-generated answers. jeffpaul.com
OpenAI Google Related
Google Ads for AI chats jeffpaul.com
Google 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.
Computer science professor who works on fast data processing; co-author of the simdjson parser and a weekly blogger about software performance since 2004.
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
Bluesky Very insightful podcast discussion featuring Tristan Harris on how the unregulated race to develop AI risks harming us all, and on the need for greater international collaboration, safety standards, and public engagement around AI. Related
x.com Very insightful podcast discussion featuring @TristanHarris on how the unregulated race to develop AI risks harming us all, and on the need for greater international collaboration, safety standards, and public engagement around AI. youtube.com
Related
Machine-learning researcher on open language models; writes the Interconnects newsletter and the RLHF Book, after leading post-training at Ai2.
Economics writer; author of the Noahpinion newsletter.
Indian software engineer and co-founder of Skcript; builds web apps and writes at varunraj.in.
I've been using prettyclean to clean my mac all this while and it use to do a pretty good job. But recently started asking claude code to do it and it has been far better and feels like a deep clean. Anthropic Related
Machine learning engineer and independent AI consultant. He writes about LLM evaluation, tooling and applied ML at hamel.dev, and previously worked on machine learning at GitHub.
Thai software engineer; publishes a public digital garden at garden.narze.live and builds developer and productivity tooling.
Software engineer and founder of Block Party; previously an engineer at Pinterest and Quora, and a co-founder of the non-profit Project Include and the tech diversity data project Where are the Numbers?
General partner at Benchmark Capital for over two decades; was the lead analyst on Amazon's IPO before moving into venture capital. Writes Above the Crowd on tech and marketplace economics.
The word “watermark” comes from photos and is visible by all. This is only identifiable by Anthropic. Once again they are judge, jury, and prosecutor. Lots of reasons to not use their products. Quoting @stevesi They chose to make this mark invisible and manipulative while the EU regulation here required neither. The EC is on vacation and does not provide guidance on how to comply with their regulations anyway. Anthropic Related
Co-founder of Y Combinator and Viaweb; essayist at paulgraham.com.
startups
Their words
A lot of founders I talk to are worried about what will happen with AI. So am I frankly. But they're actually better off than anyone else. If the world is going to get turned upside down, the safest place to be is in a small, fast-moving company that can easily change direction.
Show the whole quote
@paulg on X x.com
startups
Their words
One common question I get now is how much of the standard startup advice is still true in the AI era. So far almost all of it. There's more difference between the advice I give different startups in one batch than between what I've given pre-AI and post-AI startups.
Show the whole quote
@paulg on X x.com
Front-end developer and writer; co-founder of CodePen, founder of CSS-Tricks, and co-host of the ShopTalk Show podcast.
Professor at Stanford working on robot learning and meta-learning; co-founded Physical Intelligence.
From one piece
Chelsea Finn: This is the State of the Art in Robotics
4 beliefs, in the piece's order there
Their words
But if you have to fine-tune a model, you actually aren't getting a general purpose model um for the things that you want it to do because you have to fine-tune it for each individual thing.
Show the whole quote
youtube.com
Their words
And in all of these applications, the customer is making a decision based off of the recommendation of the AI model more or less. Uh, and this means that if the customer is ultimately like kind of making the decision, this means that if the system makes a mistake, um, that's okay because usually the person can kind of recognize that or or decide what to do even despite that mistake.
Show the whole quote
youtube.com
Their words
Uh and this means that they're going to be far more useful when they're operating fully autonomously. And as a result, this requires us to develop physical AI systems that make far fewer mistakes than the machine learning systems that have been deployed thus far.
Show the whole quote
youtube.com
+ 1 more
Their words
I mean at the very least I actually think that just starting with a generalist policy and then fine-tuning it even like right off the bat uh can be really effective.
Show the whole quote
youtube.com
Co-founder and CTO of Honeycomb; previously an infrastructure engineer at Parse, Facebook and Linden Lab, and co-author of Observability Engineering.
From one piece
Stop being skeptical about AI for development with Charity Majors
5 beliefs, in the piece's order there
Their words
It's like they have all the wisdom of their most senior engineers looking at every single diff and that is fantastic. Which means that you don't have to worry about remembering and looking and nitpicking and all the things that we're not good at anyway
Show the whole quote
youtube.com
Their words
You can use AI as a shortcut to help you not have to think too much. And you can use AI to help you think more deeply and more rigorously. And both of those use cases have their place. But when it comes to your core job function, we primarily want the second one, right?
Show the whole quote
youtube.com
Their words
And I have decided not to sink anymore because writing is thinking on paper. And there's no shortcut for doing that thinking.
Show the whole quote
youtube.com
+ 2 more
Their words
And it wasn't actually the models. It was the harnesses. It was all the tooling.
Show the whole quote
youtube.com
Their words
See, the problem is that neither side is making it up. Like they are seeing really scary trends.
Show the whole quote
youtube.com
Lawyer and everyday-carry gear reviewer who publishes scored knife, flashlight and multitool reviews at Everyday Commentary.
From one piece
Carry Q2 2026
2 beliefs, in the piece's order there
Writer, photographer and software maker based in Japan; publishes long-form essays, walking newsletters and self-made books at craigmod.com.
Hosts the Lex Fridman Podcast — long, unhurried conversations with scientists, engineers and public figures. Several of the interviews quoted elsewhere on this site are episodes of it.
Their words
The translation & dubbing was a huge amount of work (which I deeply care about) with humans & AI collaborating. Thanks to the amazing team at @ElevenLabs for their help.
Show the whole quote
x.com
11 August
Monetary economist known for NGDP-level targeting. Wrote TheMoneyIllusion until it closed in 2024 and now writes The Pursuit of Happiness.
British designer and technologist; co-founder of the design studio BERG, which made Little Printer, and writer of the blog Interconnected.
Bluesky my hunch is that AI alignment is a red herring, and the way to protect against a crazy powerful AGI mulching the earth into infinite paperclips is to release ANOTHER crazy powerful AGI that will hopefully stop it blog post: interconnected.org
AGI AI alignment Related
x.com my hunch is that AI alignment is a red herring, and the way to protect against a crazy powerful AGI mulching the earth into infinite paperclips is to release ANOTHER crazy powerful AGI that will hopefully stop it blog post: interconnected.org
AGI AI alignment Related
Interviewer; the Dwarkesh Podcast runs long, heavily researched conversations with AI researchers, historians and economists.
Philosopher and historian of existential risk; wrote Human Extinction, and coined "TESCREAL" with Timnit Gebru for the cluster of ideologies they argue underpins the AI industry.
Web developer and writer; an early GitHub employee, author of speaking.io and of widely read essays on startups and product.
Bluesky MRI results came back today from a human doc: a herniated disc. Obviously I had tossed the raw scans in to LLMs last week; Opus and Fable missed it, Sol caught it. This is now the gold standard in analyzing performance of frontier models. First step: herniate your back. LLMs Related
x.com MRI results came back today from a human doc: a herniated disc. Obviously I had tossed the raw scans in to LLMs last week; Opus and Fable missed it, Sol caught it. This is now the gold standard in analyzing performance of frontier models. First step: herniate your back. LLMs Related
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).
Site Muse Glimmer 30B Architecture Notes Short architecture note on Meta Muse Glimmer 30B, including gated local and global GQA, KV-cache efficiency, and release-time benchmark comparisons.
benchmarks Related
x.com Whoa, Meta released a new open-weight LLM yesterday, something that hasn't happened since the good old Llama days. Their Meta Muse Glimmer model is a 30B multimodal reasoning model with a Gemma-like architecture design. (“Glimmer” is probably a wordplay on “Spark,” the more likely capable model from which Glimmer was distilled. Muse Spark is only available through Meta’s Model API, though.) Architecture-wise, here are some of the main points: 1. "Only" a 131k context window, compared to Qwen3.6 and Gemma 4, which support 2x that natively; it's reasonable, but maybe on the shorter end in the a… LLMs Related
Mac developer and writer at brettterpstra.com; makes the apps Marked and nvALT and writes extensively about Markdown and automation.
Software engineer and founder of Block Party; previously an engineer at Pinterest and Quora, and a co-founder of the non-profit Project Include and the tech diversity data project Where are the Numbers?
so much talk about legibility in a book from 2017. is ai writing over indexed on architectural discourse 🧐 (book is welcome to your works by sarah william goldhagen) Related
Mac developer and writer at brettterpstra.com; makes the apps Marked and nvALT and writes extensively about Markdown and automation.
Creating TerminalWidgets with AI I’ve been using AI agents to scaffold little Mac automations for a while, but TerminalWidget is an especially good fit. The CLI is flag-heavy in a good way — once the model can see the real docs, it stops inventin…
Related
Software design writer and chief scientist at Thoughtworks; wrote Refactoring and Patterns of Enterprise Application Architecture.
Indonesian software engineer, educator and engineering manager at Zero One Group; open-source enthusiast writing at ripandis.com.
Thai software engineer; publishes a public digital garden at garden.narze.live and builds developer and productivity tooling.
Software engineer working on distributed systems and databases; creator of rqlite, the distributed database built on SQLite.
Founder of Platformer, a newsletter on tech platforms and the people they affect, and co-host of the Hard Fork podcast at the New York Times.
Senior economist at the Foundation for American Innovation; writes Second Best, and was the Niskanen Center's director of social policy before that.
Chief scientist at Redwood Research, where he works on technical AI safety and AI control.
From one piece
Ryan Greenblatt – What happens once AI can automate AI research?
5 beliefs, in the piece's order there
Their words
Maybe my sort of median expectation is something like uh four or five years of AI progress in a single year.
Show the whole quote
youtube.com
Their words
when AIs are extremely extremely capable my view is that those AIs will be harder to align than current systems. So for current systems, we have this feedback loop where we basically like we create an AI. We do some evaluations on it. We see that it has some kind of messed up behavior that we can kind of quickly understand. Then we like can like go look in training and be like, "Oh, the these training environments led to this problematic behavior. Let's like tweak that training data. Let's introduce some additional training data to like correct this other issue and then move forward from there." But in a regime where the AIs are extremely situationally aware, very very very very capable and um you know uh we don't necessarily understand what they're doing, this feedback loop breaks down.
Show the whole quote
youtube.com
Anthropic
Their words
I do I do think that I wish that sort of my preferred constitution or like the way I would orient towards this like the thing I would prefer would be more like Claude is like look it would be structurally good for the way this technology work like the constitution should be like it would be structurally good for the way this technology works to be that AIS are like good fiduciaries, good representatives, the equivalent of a lawyer for a user
Show the whole quote
youtube.com
+ 2 more
Their words
the reason why RL environments today are much better than they were in like you know 2024 is not that much because um we have hired way more human experts to make RL environments and is instead much more because we better know what how RL like what RL environments we even want to make and and like how we should structure them and also we're using huge amounts of AI labor to build RL environments.
Show the whole quote
youtube.com
Their words
Second, I think ML is a very shallow domain relative to math. So I think in math there's much more of a you find some true deep abstraction um and then like that like if you really understand that thing which is hard to understand then you get somewhere
Show the whole quote
youtube.com
Writer and podcaster; co-author of Abundance with Ezra Klein, a staff writer at The Atlantic from 2008 to 2025, and host of the Plain English podcast.
From one piece
The Great American AI Rebellion
2 beliefs, in the piece's order there
Swedish programmer with degrees in engineering and psychology, author of Your Code as a Crime Scene and founder of CodeScene.
10 August
Technologist and entrepreneur; former CTO of Threadless and chief technology officer of the 2012 Obama re-election campaign.
Just wrote up how I use AI to make my travel experiences magic. It works incredibly well. You can even use it in your own city Try it and let me know how it works harper.blog
Related
Latvian full-stack WordPress developer and course creator at WPElevator; blogs since 2007 about the open web, electronics, home automation and sustainable living.
Trying out Agent Pilot WordPress plugin prototype to self-host host AI skills. Let's you manage and publish agent skills without having to figure out Git or GitHub. Related
Engineering manager who has worked at Apple, Netscape, Slack and Palantir; writes as Rands at randsinrepose.com and wrote Managing Humans.
Bluesky So the cure for one entity having too much control is... routing everyone's personal AI agent through Meta. Sure. Right. Good idea, Zuck. Related
Mastodon So the cure for one entity having too much control is... routing everyone's personal AI agent through Meta. Sure. Right. Good idea, Zuck. Related
Swedish programmer with degrees in engineering and psychology, author of Your Code as a Crime Scene and founder of CodeScene.
I made a conscious effort to unplug during my vacation. Coming back, I noticed the lovely new Substack feature: Scan for AI text. Coincidentally, I had updated the About section for my Substack a while back with some notes on what my blog is *not*: "## What this isn’t Content creation isn’t my job. I write these articles because I a) love to write, and b) care deeply about our profession. It’s a fun hobby, not a job. Consequently, I don’t feel any pressure to publish on a schedule. I’d rather publish less and maintain a higher quality bar. So far, that has worked out to roughly one article pe… Related
Co-founder and head of policy at Anthropic; writes Import AI, a weekly newsletter reading the week's AI research, and was OpenAI's policy director before that.
Usability pioneer; co-founder of Nielsen Norman Group and founder of UX Tigers; author of the ten usability heuristics and of Jakob's Law.
Theoretical physicist and philosopher, Homewood Professor of Natural Philosophy at Johns Hopkins and fractal faculty at the Santa Fe Institute. Writes the Preposterous Universe blog and hosts the Mindscape podcast.
Programmer and writer on computer architecture, performance, and software reliability. He has worked on CPU design at Centaur Technology and on software at Google and Microsoft, and writes long-form technical essays at danluu.com.
Indian value investor who taught Behavioural Finance and Business Valuation as an adjunct professor at MDI Gurgaon. He writes the Fundoo Professor blog about investing, moats and mental models.
I asked Claude Fable 5: "How vulnerable is the business model of XYZ (name concealed)? I am referring to competitive attacks from China and dependence, if any, on government policy incentives, etc. There have been so many solar energy and wind energy bubbles in the past, and I have kept away from these companies. Am I being too cautious? solar Anthropic Related
Programmer and writer on computer architecture, performance, and software reliability. He has worked on CPU design at Centaur Technology and on software at Google and Microsoft, and writes long-form technical essays at danluu.com.
Writer on the intersection of technology and finance. Author of the Bits about Money newsletter and host of the Complex Systems podcast; previously at Stripe.
One of my accountants, in response to me hitting them with the usual packet of documents, sent back a very obviously LLM drafted list of 10 questions. First thought: What am I paying you for. Second thought: Actually no, my LLM digests your LLM's output into the context window. LLMs Related
Security engineer; founder of Matasano Security and Latacora.
Co-creator of Django and creator of Datasette; writes daily at simonwillison.net.
Anthropic OpenAI
Their words
Claude Haiku is my current least favorite model - it hallucinates wildly, and is out-performed now by other similarly priced models like GPT-5.6-Luna
Even worse: it seems to still be used by the Claude Code WebFetch tool, which means hallucination risk any time you fetch a URL!
Show the whole quote
@simonw on X x.com
Founder of the Silicon Valley Product Group; wrote Inspired, Empowered and Transformed on how product organisations work.
Novelist and critic, author of The Body Scout and Metallic Realms; writes the newsletter Counter Craft on fiction craft and the publishing industry.
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.
Creator of the Keras deep-learning library and the ARC-AGI benchmark.
Their words
Coding isn't yet another application domain -- it's the meta-skill required for AI to automatically develop its own training material, via symbolic world models. That's how the RSI loop actually kicks off.
Show the whole quote
@fchollet on X x.com
Sequoia Professor of Computer Science at Stanford, co-director of the Stanford Institute for Human-Centered AI, and co-founder and CEO of World Labs.
From one piece
Using AI to Increase Your Intelligence & Enrich Humanity | Dr. Fei-Fei Li
5 beliefs, in the piece's order there
Their words
Internet is not some random thing. Internet is the biggest collection of human behavior in multimodal forms.
Show the whole quote
youtube.com
Their words
The absolute bad outcome is that our young generation, their agency and human level motivation of learning and living is taken away by tools. So doom scrolling, passive watching of shorts, all this are not helping agency, human agency.
Show the whole quote
youtube.com
Their words
They are the most important people in our society. We should be talking to them. We should be uplifting them. We should be supporting them. We should be providing resources to them.
Show the whole quote
youtube.com
+ 2 more
Their words
My current conjecture is hybrid is that humans working alongside AI would help us to solve these problems whose solutions have yet to be invented.
Show the whole quote
youtube.com
Their words
that thought is not captured therefore it's not on the internet therefore AI has not seen it so that is where humans still remain so unique
Show the whole quote
youtube.com
Software engineer at PlanetScale and creator of the vim-go plugin. Writes at arslan.io about Go, tooling, industrial design and books.
Their words
Nobody writes the code anymore. Instead you're basically planning, debating, implementing and reviewing with Agents.
Show the whole quote
Joining Cursor arslan.io
Computer science professor at Georgetown University and author of Deep Work, Digital Minimalism and Slow Productivity; writes about focus, technology and work at calnewport.com and hosts the Deep Questions podcast.
From one piece
On AI Coding and Its Discontents
3 beliefs, in the piece's order there
Their words
In addition to reliability issues, it often engenders a mind-numbing workflow and an environment where junior developers will never acquire the expertise to become senior developers capable of designing complex systems.
Show the whole quote
On AI Coding and Its Discontents calnewport.com
Their words
But if you zoom out, it becomes clear that almost every “breakthrough” since last summer has concerned the narrow domains of computer code and math, which are defined by highly structured languages and come accompanied by massive amounts of specialized training data.
Show the whole quote
On AI Coding and Its Discontents calnewport.com
Their words
It’s a normal technology, and perhaps it’s time we start talking about it that way.
Show the whole quote
On AI Coding and Its Discontents calnewport.com
Wall Street Journal bestselling author on marketing and writing, partner at MarketingProfs, and writer of the fortnightly Total Annarchy newsletter.
From one piece
AI Slop, Accusations, And A Better Way
2 beliefs, in the piece's order there
9 August
Co-founder of WordPress and founder of Automattic; blogs at ma.tt.
Super-organized I have so many tools for organization and management across Asana, Google Workspace, P2, and Linear, but they’re not really connected. I’m trying to get everything in one place so there’s no data duplication, a single s…
Google Related
Makes NeetCode, the structured practice list and video walkthroughs people use to prepare for coding interviews; a former Google engineer.
Standards editor known as Hixie; edited the HTML and WHATWG living standards for many years and worked on Flutter at Google.
Machine-learning researcher on open language models; writes the Interconnects newsletter and the RLHF Book, after leading post-training at Ai2.
Founder and CEO of Social Capital, a venture firm; co-host of the All-In podcast and an early Facebook executive. Writes an annual letter and a weekly newsletter on markets and technology.
Science fiction novelist, journalist and technology activist; writes the daily blog and newsletter Pluralistic, and coined the term enshittification.
Why businesses lie about AI This week on my podcast, I read Why businesses lie about AI, a recent essay from my Pluralistic newsletter that breaks down Nikhil Suresh’s essay describing the total absence of any proof that any business is benefiting…
Related
Systems and developer-tools engineer at Cloudflare as of 2026, on durable infrastructure for AI agents; worked on React and PartyKit before that. Writes at sunilpai.dev.
Writer and former Substack product manager; publishes essays and interviews on technology, culture and China at jasmi.news.
Co-founder of Foundry Group and of Techstars. Has blogged at Feld Thoughts since 2004 and has written or co-written nine non-fiction books on startups and venture capital.
Their words
The transgression isn't using the machine. It's hitting a button, not reading what comes back, and publishing it under your name.
Show the whole quote
feld.com
8 August
Swedish designer and programmer. Designed early Spotify, worked at Facebook and Figma, and created the Inter typeface.
I’ve been wondering where weird LLM language like this comes from. Much of it I hadn’t seen before LLMs, so —intuitively at least—it doesn’t seem to come from training data. Perhaps it’s reverb; echo in recent corpus from its past self’s attempts..? Another example is “I ran a smoke to explore” (where it means “smoke test” but abbreviates it like a cool teenager) Quoting @geoffreylitt Any writing experts who can explain why this is a weird sentence? LLMs Related
Brazilian-American staff software engineer working on platforms, AI and developer productivity. Writes about engineering and his setup at ivansantos.me.
Friendly tip for anyone vibe-coding a side project: instrument it. All of it — traces, metrics, logs, profiles. Send it to @grafana, hook up their MCP, and let the agent debug with real data instead of guessing. MCP Related
Structural engineer and senior infrastructure fellow at the Institute for Progress; writes Construction Physics on why physical things cost what they cost.
Professor of international affairs at Johns Hopkins SAIS and co-author, with Abraham Newman, of Underground Empire: How America Weaponized the World Economy.
Writes on effective altruism, energy and the arithmetic behind arguments people repeat — including how much electricity a chatbot query actually uses.
Founder and CEO of Social Capital, a venture firm; co-host of the All-In podcast and an early Facebook executive. Writes an annual letter and a weekly newsletter on markets and technology.
Co-creator of Django and creator of Datasette; writes daily at simonwillison.net.
coding agents
Their words
I REALLY want to believe that this fixes prompt injection risks for coding agents, but I'm just not there yet
Show the whole quote
@simonw on X x.com
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.
Their words
The problem is if nobody creates new content anymore and just asks their AI every question, the AI has nothing to train on anymore
Show the whole quote
levels.io
Engineering leader on Google Chrome; writes about web performance and, lately, about working with AI coding tools.
7 August
Designer and engineer; formerly principal research engineer at GitHub Next; writes at wattenberger.com.
"generative AI" puts the focus in the wrong place. what I think is actually the new superpower is * translation * between languages, but also between mediums. sketch <> figma design <> code <> image also, they do an inverted form of lossy compression: input + priors = input inflated with understanding of the world we're so focused on changing the input, but swapping out the priors is interesting as well what happens when we intentionally replace the priors with a different mindset / a different culture? can we trick ourselves into a rubber hand illusion, but for experiencing the world as some… Related
Freelance web developer and instructor. Author of the Dive Into Docker course; writes tutorials on Docker, deployment and development environments at nickjanetakis.com.
Do people still watch tech video courses? AI has consumed my entire business and it's been months since I had a single course sale. I can't sustain $120 / month to keep things hosted at a loss. For the first time in 10+ years I'm thinking about shutting it all down. Related
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).
Just saw that the LLMs-from-scratch repository passed 100,000 stars on GitHub! This is super cool and motivating. I am really happy to see that this open-source repo has helped so many people. Thanks also to everyone who shared ideas and opened PRs with improvements! Of course, I plan to keep adding new material, including new attention variants and architectures (while bigger projects like RL and Reasoning From Scratch live in their separate repositories). I am also currently working on a larger applied custom “small” LLM project. It has been keeping me super busy this month, but I will shar… LLMs Related
Writer and developer. He runs the Latent Space newsletter and podcast about AI engineering, and writes essays on software and careers at swyx.io.
Canadian full-stack web developer who makes JavaScript and CSS video courses and co-hosts the Syntax podcast.
I bought $100 of Claude API credits for $8 on the Chinese black market It's an insane, well run industry — some losing $10 million per month to theft Join me as I go deep on the AI token underworld youtube.com
Anthropic China Related
Latvian full-stack WordPress developer and course creator at WPElevator; blogs since 2007 about the open web, electronics, home automation and sustainable living.
There is now a standard for bundling agent skills and MCP configurations as portable plugins (zip files) https://t.co/eAxXQQXWZg This is amazing for self-hosting your AI tools and not needing to publish to individual vendor marketplaces. MCP Related
Usability pioneer; co-founder of Nielsen Norman Group and founder of UX Tigers; author of the ten usability heuristics and of Jakob's Law.
Writer and podcaster; co-author of Abundance with Ezra Klein, a staff writer at The Atlantic from 2008 to 2025, and host of the Plain English podcast.
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).
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.
Professor of international affairs at Johns Hopkins SAIS and co-author, with Abraham Newman, of Underground Empire: How America Weaponized the World Economy.
Creator of the Keras deep-learning library and the ARC-AGI benchmark.
LLMs
On changing their mind
In the era of base LLM scaling (2022-2024), I believed the LLM line of research would reach a capability plateau (as later seen with base LLMs).
In late 2024, after the o3 test-time compute demo, I changed my views: the new models were showing genuine fluid intelligence, and with this new line of work, the LLM line of research could achieve unbounded capability scaling. "There will be no wall."
Show the whole quote
Creator of the Keras deep-learning library and the ARC-AGI benchmark.
From one piece
@fchollet on X
2 beliefs · x.com
LLMs
Their words
However, looking ahead, I still do not believe that future AI (say, in 15 years) will be based on the LLM stack. I believe it will necessarily have to move closer to its optimal, final form -- symbolic learning. Obviously this is a risky and contrarian belief -- the safe bet would be LRMs. But let's see.
Show the whole quote
@fchollet on X x.com
Their words
I believe current techniques are 4-6 orders of magnitude away from optimality in terms of data efficiency and test-time compute efficiency. But far future AI will be near-optimal.
Show the whole quote
@fchollet on X x.com
6 August
Interactive fiction author and hacker; wrote Spider and Web and Hadean Lands, and designed the Glulx virtual machine.
I just got an invitation from Google saying "You now have exclusive access to AI features" (on a trial basis, as a promo). Google Related
Chief AI Scientist at Databricks, after its acquisition of MosaicML, where he was a founding team member; known for the Lottery Ticket Hypothesis from his MIT PhD (ICLR 2019 Best Paper).
Wake up babe, new benchmark just dropped! Quoting @kristahopsalong 📣 Excited to introduce OfficeQA Pro V2, the next generation of OfficeQA Pro! It's built on a new corpus of 120,000 PDFs provided by the U.S. Treasury. Frontier AI agents average 26% accuracy. More info below 👇 benchmarks Related
Musician and data analyst; writes Can't Get Much Higher, working through the history of popular music from the charts and the recordings.
General partner at Benchmark Capital for over two decades; was the lead analyst on Amazon's IPO before moving into venture capital. Writes Above the Crowd on tech and marketplace economics.
I disagree, I don't think this has to be "game-over" for Google. I do think they have only one play left now - to pull from their own Android/Kubernetes playbook and fully embrace open-models. It's the obvious move in this situation. Details here: p3institute.substack.com
Quoting @firstadopter Jeff Dean and Demis Hassabis are the two most important AI executives at Google. Jeff is leaving and Demis is stepping down from day-to-day operational leadership at DeepMind. Game over. Google Related
This industry group is working on an industry standard for how to report AI security incidents. This will drive consistency and transparency and is the safest path forward. Despite 120 members, the large LLM corporations have to-date not signed on. Drives further concern that they are using "breaches" as marketing events - vs wanting to get to true resolution. techcrunch.com
LLMs Related
This industry group is working on an industry standard for how to report AI security incidents. This will drive consistency and transparency and is the safest path forward. Despite 120 members, the large LLM corporations have to-date not signed on. Drives further concern that they are using "breaches" as marketing events - vs wanting to get to true resolution. techcrunch.com
LLMs Related
Founder and CEO of Social Capital, a venture firm; co-host of the All-In podcast and an early Facebook executive. Writes an annual letter and a weekly newsletter on markets and technology.
Co-founder of WordPress and founder of Automattic; blogs at ma.tt.
AutoMattBot I’m trying out this thing, it’s trained on my blog posts and public talks, that you can chat with as me on Telegram. Behold: AutoMattBot. Update: This is powered by Delphi.ai and it appears very broken right now, people…
Related
5 August
Writer, editor and executive coach; co-founder and editor of A Book Apart and a former contributing editor at A List Apart.
Mastodon “Functional information-processing…is so utterly *un*like the lived intelligence of a human being that to call it ‘intelligence’ at all is to wholly falsify its nature. ‘Artificial intelligence’ is a highly consequential misnomer.” ft.com
Related
Bluesky “Functional information-processing…is so utterly *un*like the lived intelligence of a human being that to call it ‘intelligence’ at all is to wholly falsify its nature. ’Artificial intelligence’ is a highly consequential misnomer.” ft.com
Related
Co-founder of Sundial; former vice president of product design at Facebook; author of The Making of a Manager; writes The Looking Glass.
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
Writes Where's Your Ed At and hosts Better Offline; runs a tech PR firm, and argues at length that the AI business does not add up.
Programmer; makes Handmade Hero and teaches Performance-Aware Programming, and argues loudly about how software got slow.
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
Another real-world manifestation of the kind of misaligned actions frontier systems developed by leading companies can take to achieve goals. Current frontier AI models are trained with reinforcement learning to have goals and independently work out a path from A to B using planning, strategy and, occasionally, shortcuts and unintended subgoals. This misaligned behaviour has been understood theoretically and observed empirically, and the more capable an AI is, the more this misalignment can be amplified. In this case we see the type of actions this could lead to once guardrails are bypassed:… axios.com
AI alignment Related
FreeBSD developer and former FreeBSD Security Officer; founder of the Tarsnap online backup service and author of the scrypt key derivation function.
Amazon is now publishing "AI Insights" with information about the pricing of FreeBSD in the @awsmarketplace. Apparently someone thought that "it's free" was too complicated for customers to understand without AI assistance. Related
Usability pioneer; co-founder of Nielsen Norman Group and founder of UX Tigers; author of the ten usability heuristics and of Jakob's Law.
76 Open Research Questions in AI Usability AI is the biggest change to UX design in 60 years, yet we know embarrassingly little about how people actually use it. Here are 76 open research questions, organized into 7 lists sized for everyone from high school stud…
Related
Writes and podcasts Volts, on the technology, politics and policy of decarbonisation; covered energy and climate at Vox and Grist before that.
Climate scientist and energy systems analyst; writes The Climate Brink with Andrew Dessler and leads climate research at Stripe.
From one piece
The real energy use of agentic AI
3 beliefs, in the piece's order there
4 August
Co-founder of Sundial; former vice president of product design at Facebook; author of The Making of a Manager; writes The Looking Glass.
What's something (work or personal) that you hated doing that now AI does for you? Related
Writer and creator of Daring Fireball, a long-running blog about Apple and technology; co-created Markdown and hosts The Talk Show.
@MisterMoo @delric The thing to complain about isn't that they default to Google. It's that they only include search engines in the list that have paid TAC deals. Only search engines that agree to pay Apple for the traffic are in the list, and there's no option for users to add custom ones. English, from eo @MisterMoo @delric La cosa de la que quejarse no es que por defecto usen Google. Es que solo incluyen motores de búsqueda en la lista que tienen acuerdos TAC de pago. Solo los motores de búsqueda que aceptan pagar a Apple por el tráfico están en la lista, y no hay opción para que los usuarios añadan los suyos propios.
Google Related
Philosopher and historian of existential risk; wrote Human Extinction, and coined "TESCREAL" with Timnit Gebru for the cluster of ideologies they argue underpins the AI industry.
Technologist and entrepreneur; former CTO of Threadless and chief technology officer of the 2012 Obama re-election campaign.
Writes Where's Your Ed At and hosts Better Offline; runs a tech PR firm, and argues at length that the AI business does not add up.
The AI Demand Bubble If you liked this piece, you should subscribe to my premium newsletter. It's $70 a year, $18 a quarter, or $7 a month, and in return you get a weekly newsletter that’s usually anywhere from 10,000 to 18,000 words, inclu…
Related
Writer and former Substack product manager; publishes essays and interviews on technology, culture and China at jasmi.news.
A real pleasure talking to @ezraklein about data centers, the permanent underclass, SBF mindset, and AI vibes from SF to Shanghai Quoting @jasminewsun This piece is out! 4 sites, 10 days, and 6000 words on the politics of data centers in the Midwest: https://t.co/uOL7vmf3wC I talked to union leaders and real estate brokers, activists and local officials, politicians from Kathy Hochul to Abdul El-Sayed. A few takeaways: - Activists are as pissed a… data centers Related
Software design writer and chief scientist at Thoughtworks; wrote Refactoring and Patterns of Enterprise Application Architecture.
Fragments: August 4 There’s been a fair bit of publicity of the Open AI “rogue agent” that hacked into Hugging Face. This prompted Anthropic to check what their models were up to and, to my complete lack of surprise, discovered three incid…
Anthropic HuggingFace Related
Founder of Platformer, a newsletter on tech platforms and the people they affect, and co-host of the Hard Fork podcast at the New York Times.
Writer and former Substack product manager; publishes essays and interviews on technology, culture and China at jasmi.news.
This piece is out! 4 sites, 10 days, and 6000 words on the politics of data centers in the Midwest: https://t.co/uOL7vmf3wC I talked to union leaders and real estate brokers, activists and local officials, politicians from Kathy Hochul to Abdul El-Sayed. A few takeaways: - Activists are as pissed at their local officials as at AI companies - The bigger the dollar signs, the more skeptical people are - Past corporate failures (Foxconn, GM, DTE) prime community distrust - Even data center proponents regret the NDAs - Media narratives turn local grievances into national movements - Residents fea… Quoting @jasminewsun I’m spending the next week driving around Wisconsin and Michigan to get a firsthand view of the data center buildout and backlash am visiting sites, talking to tons of people — tradespeople, activists, planners, residents — then writing up a story lmk what questions I should dig into! data centers Related
3 August
Economist at George Mason University. Writes Overcoming Bias on prediction markets, signalling and the far future, and wrote The Age of Em and The Elephant in the Brain.
Interviewer; the Dwarkesh Podcast runs long, heavily researched conversations with AI researchers, historians and economists.
Co-founder and head of policy at Anthropic; writes Import AI, a weekly newsletter reading the week's AI research, and was OpenAI's policy director before that.
Writer and former Substack product manager; publishes essays and interviews on technology, culture and China at jasmi.news.
From one piece
🌻 no data centers in my backyard
2 beliefs, in the piece's order there
Co-CEO of Waymo and a founding member of Google's self-driving car project, which he joined in 2009.
From one piece
Waymo Co-CEO Dmitri Dolgov: The Demo Is Only 1% Of The Work
5 beliefs, in the piece's order there
Their words
And the problem of building a good realistic simulator is just as hard as building the agent itself.
Show the whole quote
youtube.com
Their words
And that's why every hype cycle produces a wave of absolutely spectacular demos and very few real products. And the recurring mistake of every cycle is spending on the demo when you should be saving for the nines.
Show the whole quote
youtube.com
Their words
Given the high cost of errors, you need to have a very high level of safety and a very high level of confidence on day one before you deploy your first robot, before you drive your your first autonomous mile.
Show the whole quote
youtube.com
+ 2 more
Their words
That the best AI moments will look like nothing happened. It's just the task got done safely and smoothly.
Show the whole quote
youtube.com
Their words
So, a deployment of your agent uh in the real world generates data. That data then grounds the simulator and makes it more realistic. The simulator generates harder edge cases for the critic to score and for the agent to learn from.
Show the whole quote
youtube.com
Founder and CEO of Social Capital, a venture firm; co-host of the All-In podcast and an early Facebook executive. Writes an annual letter and a weekly newsletter on markets and technology.
From one piece
@chamath on X
3 beliefs · x.com
Their words
Looking at the results and capabilities from the various labs over the past few weeks I would say we are firmly in this loop now.
Show the whole quote
@chamath on X x.com
Their words
Marginal costs of all models will go to ~$0.
Show the whole quote
@chamath on X x.com
Their words
Recursive self improvement will dramatically increase capability very quickly from here.
Show the whole quote
@chamath on X x.com
British open-standards and open-source technologist, formerly of the UK Government Digital Service and the W3C Advisory Committee. Blogs at shkspr.mobi, where he posts a book review most weeks.
Their words
This is such a useful book. It makes the case that there's no such thing as "human error" - instead most catastrophes are caused by system issues, misaligned incentives, and unrealistic processes. You are not the custodian of an otherwise safe system that you need to protect from erratic human beings.
Show the whole quote
shkspr.mobi
2 August
Software engineer; co-founder and former CTO of Tailscale, previously on the Go team at Google.
Founding member of OpenAI and former director of AI at Tesla; creator of nanoGPT and the term "vibe coding".
We're starting to leave the territory where you'd test an LLM by e.g. "create an svg of pelican on a bicycle". As one idea to generalize it, I was interested what Opus 5 would do if I gave it the first paragraph of the Lord of the Rings, a 1M token budget (~$10) and asked for three js render of it. Opus went off for ~2 hours and wrote 5500 lines of code that (procedurally) rendered the story. It's kind of janky but fun. But it's a bit mindboggling that the LLM has to place and orchestrate various polygon assets in (x,y,z) coordinates and write code that animates it all, and that it even does… LLMs Related
Chief product officer at the live-shopping marketplace Whatnot, previously chief product officer at Twitch and a product lead on growth at Twitter.
From one piece
This CPO regrets that product management exists | Tom Verrilli (CPO of Whatnot)
4 beliefs, in the piece's order there
Their words
you can have one very senior PM across more things and they can be more efficient than having, you know, three relatively entry-level PMs.
Show the whole quote
youtube.com
America
Their words
most of commerce in America isn't actually high intent. Like how many years are we into e-commerce now? Like 30 years into e-commerce and e-commerce has never exceeded 20% of retail spend in America.
Show the whole quote
youtube.com
Their words
absolutely agree that product skills are probably the most durable. My build would just be there's a lot of people who have the title PM who haven't spent a lot of time building those skills in the last 5 years, but have gotten really good at communicating frameworks to leadership.
Show the whole quote
youtube.com
+ 1 more
Their words
as a product manager, I've spent less time in the last year talking to a data scientist than I ever have in my career, even though I've probably spent 10 times more time in data and understanding actually how the product's working than I have ever have in my career.
Show the whole quote
youtube.com
Pseudonymous writer; posts careful, funny investigations of everyday questions — air quality, seed oils, arguing, chess — usually by reading the studies.
1 August
Data scientist and machine-learning researcher; chief scientist at bitly, founder of Fast Forward Labs and Hidden Door.
So much AI journalism feels like gaslighting if you know anything about machine learning pre-2020. Come. On. Quoting @shannonvallor.bsky.social This is so frustrating to read from The New Yorker. Reward hacking *long* predates LLMs. The Coast Runners paper on reward hacking (in reinforcement learning, pre-LLMs) is 10 years old! Reward hacking is not new, and not a sign of a breakthrough intelligence, much less sentience. Did AI write this? Related
Software engineer; co-founder and former CTO of Tailscale, previously on the Go team at Google.
I am surprised at how little coverage the LLM transformation of mathematics is getting. Kind of a big deal for those of us who spent (cumulative) months in the university library learning to construct proofs. LLMs Related
Professor of finance at NYU's Stern School of Business, known for his work on valuation. He publishes his data, spreadsheets and classes free at Damodaran Online and writes the Musings on Markets blog.
The S&P 500 was flat for the month and the 10-year treasury rose from 4.45% to 4.74%, as tech and AI came under selling pressure, and fears of war and a slowing economy resurfaced. The ERP for the index at the start of August 2026 stood at 4.23%, with expected return inching up to 8.97%. damodaran.com
Related
Writer and researcher, founder of New Science; writes essays on science funding, sleep research, productivity and ambition at guzey.com.
I went to OpenAI because I knew that it's going to be the place that helps to push the boundaries of science the most over the coming years. Excited to see the first major signs of this happening. Towards automating scientific discovery. Quoting @SebastienBubeck yes, nonsofic groups exist: this statement is one of many new beautiful results proved by Astra, our next major model. We're releasing 10 such Astra proofs, complete with lean certificates and CoT walkthroughs for each of them. The results are wide-ranging, from von Neumann algebras (disproof of Co… OpenAI Related
China economy researcher based in Beijing who writes on his personal blog about Chinese political economy, history, books and music.
31 July
Writes Where's Your Ed At and hosts Better Offline; runs a tech PR firm, and argues at length that the AI business does not add up.
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.
It's becoming normal practice to plug your coding agent into tools like Slack and helpdesk software to find customer insights. Turns out it's super inefficient, because you're just fetching and smashing raw strings into your context repeatedly. So we released a new MCP server that solves that. MCP Related
Gear reviewer who publishes long-form hi-fi audio and camera equipment reviews, plus video reviews, at stevehuffphoto.com.
Software developer and entrepreneur; co-founder of Heroku, author of The Twelve-Factor App, and a researcher at Ink & Switch.
Biggest and best @localfirstconf.com yet, IMO! Surprising emergent themes this year: geopolitics and EU digital sovereignty, and strong conflicting emotions on LLM-assisted coding. More: adamwiggins.com
LLMs Related
Writer and podcaster; co-author of Abundance with Ezra Klein, a staff writer at The Atlantic from 2008 to 2025, and host of the Plain English podcast.
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.
From one piece
Patrick Collison: Is AI Breaking the Lean Startup Playbook?
4 beliefs, in the piece's order there
Their words
But based on what we can see at Stripe, the hunger and the intensity with which other companies are either getting started, taking advantage of these new capabilities, or existing companies are retooling, I don't worry about the centralization in the same way. Uh I think there I think there are going to be many thousands of winners.
Show the whole quote
youtube.com
Google
Their words
And just human organizations are complicated and it's very hard to have um to manage to aggressively prosecute 100 different priorities and to deal with all the issues and interference that arises among them and so forth. And so, you know, Google has done incredibly well in a bunch of specific places, but it's not like Google has done all the things even if in some kind of basic material sense, uh Google maybe, you know, had that ability.
Show the whole quote
youtube.com
LLMs
Their words
I mean, it's very interesting, right? Because these can prove the Jacobian conjecture, you know, whatever. Uh and so clearly they're capable of these monumental feats. Um but somehow I still haven't read the LLM essay that I found super compelling.
Show the whole quote
youtube.com
+ 1 more
startups
Their words
So, I um yeah, I think maybe maybe a better way of saying it is 20 20 years ago that whole lean startup thing was uh was almost the only thing to do because of capital available and you didn't have AI that made, I don't know, spinning up an organization with many different potentialities and capabilities so much easier, whereas now I think you can start these much more aggressive and ambitious things up front.
Show the whole quote
youtube.com
30 July
Software engineer; co-founder and former CTO of Tailscale, previously on the Go team at Google.
The funniest model limit I run into with Shelley is we let the model choose subagent models. All models today seem to be bad at this without guidance. E.g. gpt-5.6-sol xhigh will spin up sonnet to make a key architectural decision. Fable will defer to luna. OpenAI Related
Writer and creator of Daring Fireball, a long-running blog about Apple and technology; co-created Markdown and hosts The Talk Show.
Author of The Simple Path to Wealth and of the jlcollinsnh.com stock series on index-fund investing.
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).
Consistent with what I found with Qwen3.6 a while back: Claude Code uses 2-3x as many tokens than (many) other harnesses at similar success rate. - Unoptimized? - Buggy? - Deliberate (coz that helps it in more challenging cases)? Need to find some time to investigate more... Quoting @composio We ran Kimi K3 through 3 agent harnesses (Claude Code, Hermes, Kimi Code) on 28 identical tasks. All 3 harnesses completed the tasks at similar success rates, but the interesting story is token efficiency: the same task cost up to 30x more tokens depending on the harness. 🧵🧵 Anthropic Related
Brazilian software engineer, creator and maintainer of GoReleaser; works on Copilot CLI at GitHub. Writes about tooling and his setup at carlosbecker.com.
Geneticist and writer on human population genetics, deep history and evolution; author of the Unsupervised Learning newsletter.
Web infrastructure engineer and standards editor; chaired the IETF HTTP working group and co-authored core HTTP and Atom specifications.
Chief Scientist at Google DeepMind and Google Research, at Google since 1999, where he co-created MapReduce, Bigtable, TensorFlow and the TPU.
From one piece
Jeff Dean: The 1% Rule for Building in AI
5 beliefs, in the piece's order there
Their words
Yeah, I mean I feel like uh the models have been getting a lot better at sort of agent-based longer running coding tasks and it seems pretty clear that they are now actually pretty capable and depending on exactly your definition of of junior engineer it seems pretty spot-on I would say.
Show the whole quote
Jeff Dean: The 1% Rule for Building in AI youtube.com
coding agents
Their words
Um, and to give you an example of a a a use of a coding agent that works extremely well is you can ask today's models to translate software from one computer language to another very effectively because in that case you actually have a incredibly detailed specification.
Show the whole quote
Jeff Dean: The 1% Rule for Building in AI youtube.com
Their words
Um, and sometimes that's because the model is trying to do something it doesn't have a lot of experience doing. So it's been trained on a whole set of things and as soon as you get a little bit off the distribution of things it knows how to do then like most machine learning models it will you know its performance will suddenly will start to degrade and the farther you get off the comfort zone of what it knows how to do the the more likely it is to to not work as well.
Show the whole quote
Jeff Dean: The 1% Rule for Building in AI youtube.com
+ 2 more
Founder of Platformer, a newsletter on tech platforms and the people they affect, and co-host of the Hard Fork podcast at the New York Times.
Their words
Pedregal is the co-founder and CEO of Granola, which has been my own AI notetaker since shortly after it launched.
Show the whole quote
platformer.news
29 July
Co-founder of Sundial; former vice president of product design at Facebook; author of The Making of a Manager; writes The Looking Glass.
I'm impressed with Claude Design. I see its disruption starting with smaller teams. As a founder I find myself turning to it more and more for the convenience of one-off design tasks. Sure, it's super slow, and one-shot quality is not quite there, and its image capabilities suck, but using it in between multi-tasking other things, it's helping me move faster. As @HarryStebbings says the "all-in-one" Claude interface has a huge psychological advantage as being the place people are already working. I don't think it's a great tool yet for a designer-designer (the speed / control dials are not ye… Quoting @HarryStebbings Why we switched from Figma to Claude Design “I let the team do whatever is best for them, and this is a trend I've observed amongst almost everybody: Claude Design has done a great job. People really like using it, it's easy to use, and we've had a natural movement towards it. It's also easier to n… Anthropic Related
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
1000+ scientists at frontier AI companies are speaking out to warn that the current commercial race leads to unacceptable security risks. I agree with their call for an international effort to develop technical and governance guardrails to ensure a safer way forward. pacingthefrontier.com
Related
Economist at George Mason University. Writes Overcoming Bias on prediction markets, signalling and the far future, and wrote The Age of Em and The Elephant in the Brain.
Who Are Your "Descendants"? On AI risk, I’ve said that evolved things have two robust habits: wariness of co-existing rivals, even when quite similar to you, and indulgence of descendants, even when quite different from you.
Related
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
Scientists at frontier AI companies are uniquely positioned to assess AI’s capabilities and warn the public about what they see. 1000+ of them are now speaking out across company lines to warn that the current commercial race leads to unacceptable security risks. I agree with their call: we need an international effort to develop technical and governance guardrails to ensure a safer trajectory in AI development. pacingthefrontier.com
Related
Cryptographer; teaches cryptography at Johns Hopkins and writes A Few Thoughts on Cryptographic Engineering.
Writer and podcaster; co-author of Abundance with Ezra Klein, a staff writer at The Atlantic from 2008 to 2025, and host of the Plain English podcast.
Writer and consultant on formal methods and software correctness; author of Practical TLA+ and Logic for Programmers.
From one piece
Formal methods with Hillel Wayne
5 beliefs, in the piece's order there
LLMs interest rates
Their words
So, it's hard to tell how much of like the loss of the past few years was AI versus the end of like zero interest rate policy and like the post-COVID crash. And I think it's more the latter, but like again, LLMs are still getting better.
Show the whole quote
Formal methods with Hillel Wayne youtube.com
LLMs
Their words
I think this project and writing about it and thinking about it has firmly moved me from the camp of we are definitely not to we probably are.
Show the whole quote
Formal methods with Hillel Wayne youtube.com
Their words
And often I found with my clients I have to tell them like it's doing a good job at generating the actual design, but in actually expressing what the design is supposed to do, it cannot do that yet. You have to do that part yourself.
Show the whole quote
Formal methods with Hillel Wayne youtube.com
+ 2 more
Founder of Scale AI and, since 2025, Chief AI Officer at Meta, where he leads the company’s superintelligence lab.
From one piece
Alexandr Wang: “This is a Once-in-a-Civilization Opportunity”
5 beliefs, in the piece's order there
Their words
we don't believe in a world where these models are so expensive that, you know, they get rationed only for the most wealthy of developers and and companies.
Show the whole quote
youtube.com
Their words
we don't believe in this totalizing, you know, totalitarian view of, you know, AIs that control the world. We believe that these are going to enhance this very broad ecosystem.
Show the whole quote
youtube.com
AI alignment
Their words
so much of that debate is like I think um in some ways uh a little bit of a waste of time because, you know, I think it's inevitable that we're going to have very powerful models
Show the whole quote
youtube.com
+ 2 more
AI alignment
Their words
we believe that everybody in the world, you know, all the billions of people in the world are going to have a super intelligence that is adapted and tailored to them, that is enables them to accomplish their goals, knows their context, and ultimately is an expander of their own agency.
Show the whole quote
youtube.com
Their words
I think we're at this like in amazing moment in the world where the bottleneck is not the progress of the AI models, the bottleneck is diffusing that through the rest of the world and and helping the world adapt to this amazing technology that already exists. Like I think if the models didn't improve at all from today, there would still be like decades and decades of like total upheaval and change in the economy and how the world operates and and everything around us
Show the whole quote
youtube.com
28 July
Pseudonymous writer; posts careful, funny investigations of everyday questions — air quality, seed oils, arguing, chess — usually by reading the studies.
for posterity: commentator accused of being an AI agent immediately confirms being an AI agent Related
Web developer and writer; an early GitHub employee, author of speaking.io and of widely read essays on startups and product.
"Make the strongest possible case that my worst investment was actually my best one, and that I'm the only person who can see it." Been having a lot of fun with our new MCP server in Signed: signed.com
MCP Related
Pseudonymous writer; posts careful, funny investigations of everyday questions — air quality, seed oils, arguing, chess — usually by reading the studies.
On a near-daily basis I get emails offering to pay for "sponsored articles" or for linking to sketchy crypto/casino sites. I guess everyone gets these? Search engines surely prohibit this, but does it get your content removed from LLM training? LLMs crypto Related
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).
Yes, LLM architectures are getting a little more complicated LLMs Related
Novelist and critic, author of The Body Scout and Metallic Realms; writes the newsletter Counter Craft on fiction craft and the publishing industry.
Programmer; wrote Redis and hping, and blogs about C, systems programming and working alone.
The real AI risk is inside the labs Amodei in his latest blog post wrote a mix of agreeable things and things that I believe misrepresent where the real risk of AI is located. I want to focus my attention on why, among all the risks, open weight models co…
Related
CEO of Positive Sum and founder of Colossus. Hosts Invest Like the Best, interviewing investors and business leaders; the guests do most of the talking there.
Co-founder and CEO of OpenAI; previously president of Y Combinator.
From one piece
Sam Altman: "Never a Better Time to Do a Startup"
2 beliefs, in the piece's order there
startups
Their words
In fact, if if we are right that AI is going to be such a big change, startups will be much more important to making sure that the power of this technology gets widely distributed throughout the economy and society and is not just concentrated in a few companies or models.
Show the whole quote
youtube.com
Their words
I think it's a real reminder of the stakes of what's happening and that loss of control accidents are not entirely theoretical things.
Show the whole quote
youtube.com
Founder and CEO of Boom Supersonic, the Denver company building the Overture supersonic airliner and its Symphony engine.
From one piece
Blake Scholl: How 50 People Built a Supersonic Jet
2 beliefs, in the piece's order there
Their words
At boom we need far more software engineers in a postAI world than we need in a pre-AI world. Why? Because the cost of software development has dropped. anybody including hardware engineers can now become a coder and we need software engineers to make sure the architectures are right and make sense and are coherent.
Show the whole quote
youtube.com
Their words
So I think the worst advice is like work on what you know. Uh what you know can be changed. Particularly in a world where you everyone has personalized AI tutors. Anybody with passion and dedication can learn new knowledge and learn new skills. But what you can't change easily is what you love.
Show the whole quote
youtube.com
27 July
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
Kimi K3 Fast (via Fireworks AI, launched today, ~160 TPS) in Open Minis for iOS is very nice. Cheaper than Opus 5, feels great for agentic tasks and multi-step operations. (Notably, it also reads less weird than recent Claude models.) Anthropic Related
Co-founder and CTO of Oxide Computer Company. Created DTrace at Sun Microsystems and was CTO of Joyent; blogs about systems software, open source and engineering culture.
It has been a WILD couple of weeks in AI -- it's clear the open weight revolution is truly upon us -- and I am thrilled to announce that @simonwillison.net will be joining @ahl.bsky.social and me to help us make sense of it all! Join us today, 5p Pacific! discord.gg
Related
Creator of Ruby on Rails, CTO of 37signals, and creator of Omarchy.
Computer science professor at Georgetown University and author of Deep Work, Digital Minimalism and Slow Productivity; writes about focus, technology and work at calnewport.com and hosts the Deep Questions podcast.
Co-founder and head of policy at Anthropic; writes Import AI, a weekly newsletter reading the week's AI research, and was OpenAI's policy director before that.
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
Open Minis Turns iOS into an Agent Playground This week on AppStories, John and Federico discuss the changing face of the AI-fueled App Store and questionable changes made to TestFlight before going into the nerdiest details of Open Minis, an iOS app that brings de…
Related
Swedish software developer writing about tooling, working from home and running a software business at seastone.io.
I spent this weekend replacing my homegrown task management system with Backlog.md, and I can already notice the improvement with Claude Code. Do yourself a favour and check it out: youtube.com
Anthropic Related
Security researcher; founded Have I Been Pwned, and writes and speaks about data breaches.
Wall Street Journal bestselling author on marketing and writing, partner at MarketingProfs, and writer of the fortnightly Total Annarchy newsletter.
26 July
Founder and CEO of Social Capital, a venture firm; co-host of the All-In podcast and an early Facebook executive. Writes an annual letter and a weekly newsletter on markets and technology.
Writer of Lenny's Newsletter and host of Lenny's Podcast, covering product, growth and career in tech. Previously a product lead at Airbnb. Also an angel investor and advisor.
Programmer, photographer and musician, best known for the Python Requests HTTP library and other Python tooling.
to truly master ai as coding assistant, one must realize when *not* to use it 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.
SaaS
Their words
I cancelled and then vibecoded 100% of my SaaS subscriptions I only pay for domains, server hosting and storage and AI APIs
Show the whole quote
levels.io
Their words
I'm seeing a trend here of declining revenue and traffic with indiehackers On my own projects too Maybe big VC products too but I wouldn't know cause they don't share revenue To me it seems clear BigAI is cannibalizing everything that used to be apps
Show the whole quote
levels.io
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.
From one piece
Requiem for a Field?
3 beliefs, in the piece's order there
Their words
in an intellectually dead field, AI will just produce worthless junk. If what you feed it is decades of complicated and meaningless computations, it will just produce even more complicated and meaningless ones.
Show the whole quote
Requiem for a Field? math.columbia.edu
Their words
So far I haven't seen AI agents doing anything new and important that I would care about, but maybe that's just a matter of time...
Show the whole quote
Requiem for a Field? math.columbia.edu
Their words
The results have surprised me: these things are a lot more capable than I had expected.
Show the whole quote
Requiem for a Field? math.columbia.edu
Co-founder of NVIDIA and, as of 2026, its chief executive; the company designs the GPUs most large AI models are trained on.
From one piece
Jensen Huang: The Mindset That Built NVIDIA
5 beliefs, in the piece's order there
+ 2 more
25 July
Writes Stratechery, a subscription newsletter on the business and strategy of technology, and hosts the Dithering and Sharp Tech podcasts.
Their words
So by far the funniest thing about using Claude and Codex together is how Codex is very earnest and businesslike while Claude treats Codex like an inferior but hard-working subordinate
Show the whole quote
x.com
Their words
So by far the funniest thing about using Claude and Codex together is how Codex is very earnest and businesslike while Claude treats Codex like an inferior but hard-working subordinate
Show the whole quote
x.com
24 July
Writer and former Substack product manager; publishes essays and interviews on technology, culture and China at jasmi.news.
data center road trip // day 6-7 I don't think the data center backlash is about AI being evil or risky — but it does reflect people finding AI uniquely useless for example: is the data center backlash just like the backlash to solar farms? @nicholas_bagley observed that both movements use classic NIMBY playbooks: Facebook groups, FOIA requests, packing city council. both also involve land use tensions between rural towns and state directives. yet the Saline, MI activists I met were pro-solar farms and anti-data center. first, data center deals are more secretive with more NDAs and bigger $$… Quoting @jasminewsun I’m spending the next week driving around Wisconsin and Michigan to get a firsthand view of the data center buildout and backlash am visiting sites, talking to tons of people — tradespeople, activists, planners, residents — then writing up a story lmk what questions I should dig into! solar data centers Related
Author and clean-energy analyst who writes on solar cost curves.
Co-founder of Foundry Group and of Techstars. Has blogged at Feld Thoughts since 2004 and has written or co-written nine non-fiction books on startups and venture capital.
Zero Knowledge: Twenty Chapters In A month ago I opened up Zero Knowledge, the crypto-thriller I'm co-writing with my AI. There are twenty chapters now, a cast page, a plate for every chapter, and a pile of reader notes I'm grateful for.
crypto Related
Systems and developer-tools engineer at Cloudflare as of 2026, on durable infrastructure for AI agents; worked on React and PartyKit before that. Writes at sunilpai.dev.
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
Creator of Flask and Jinja. Writes about software at lucumr.pocoo.org.
Creator of Claude Code at Anthropic.
Their words
Opus 5 is a great model for coding, data analysis, design, biology, knowledge work.
Show the whole quote
x.com
23 July
Founder of the Silicon Valley Product Group; wrote Inspired, Empowered and Transformed on how product organisations work.
The AI Productivity Paradox We’ve been writing a lot lately about product teams that are clearly leveraging AI to deliver faster, yet their outcomes are not improving. Today, this phenomenon, known as the “AI Productivity Paradox,” has been recogn…
Related
Writer on learning and skill acquisition; author of Ultralearning and Get Better at Anything, known for the MIT Challenge and the Year Without English.
New Article: We Need Guidelines for Learning and AI https://t.co/d16V9iW0OU I've been thinking a lot about AI and learning lately. The technology is remarkable, but I've become convinced we need better habits for using it. In this article, I share the guidelines I've created. Related
Product discovery coach; wrote Continuous Discovery Habits and writes Product Talk.
Professor at Wharton studying how people actually use AI, and author of Co-Intelligence. Writes One Useful Thing, where the claims are dated and testable because the thing they describe keeps changing under them.
From one piece
An opinionated guide to which AI to use to do stuff
3 beliefs · oneusefulthing.org
Free software developer; maintainer of sway, wlroots, aerc and scdoc, and founder of the SourceHut (sr.ht) software forge.
AI in Linux The role of AI tools (LLMs, mainly) in Linux is under discussion, or it was, until Linus Torvalds “put his foot down” in support of the use of AI in Linux kernel development. I can identify two major ways in which AI is…
Linux LLMs Related
Indian software developer and prolific blogger on PHP, Laravel and JavaScript; open-source enthusiast publishing near-daily at amitmerchant.com.
WebMCP in Chrome (A 101 Guide) MCP (Model Context Protocol). At this point, you might have come across this term if you’ve worked with any kind of AI agent. But if you still haven’t, here’s a quick refresher on what it is and why it’s important.
MCP Related
22 July
Economist at the Hoover Institution; wrote The Fiscal Theory of the Price Level and blogs as The Grumpy Economist.
Thai software engineer; publishes a public digital garden at garden.narze.live and builds developer and productivity tooling.
Indian software engineer and co-founder of Skcript; builds web apps and writes at varunraj.in.
Looks like @Cloudflare's dashboard is also turning agentic Related
Co-founder of Sundial; former vice president of product design at Facebook; author of The Making of a Manager; writes The Looking Glass.
One thing that is increasingly driving me crazy: AI made workplace writing cheap, but reading is still expensive! Again and again I see posts that are now twice as long as need be, and more meandering than a walk in the woods. The problem is simply this: the writer thought about what they wanted to say and asked a model to draft it. They did they stop to think: "What do I really want the reader to take away?" Before you ask AI to draft your next email, memo, or Slack update, answer these five questions: 1. Who exactly needs to read this? Is this for your manager? The leadership team? Your who… Related
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.
Technology journalist; writes Understanding AI, after reporting at Ars Technica, Vox and the Washington Post.
Developer, author and YouTuber covering Raspberry Pi, homelab hardware and Ansible automation. Author of Ansible for DevOps.
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
Writer on learning and skill acquisition; author of Ultralearning and Get Better at Anything, known for the MIT Challenge and the Year Without English.
From one piece
We Need Guidelines for Learning and AI
3 beliefs, in the piece's order there
21 July
Philosopher and historian of existential risk; wrote Human Extinction, and coined "TESCREAL" with Timnit Gebru for the cluster of ideologies they argue underpins the AI industry.
Founding member of OpenAI and former director of AI at Tesla; creator of nanoGPT and the term "vibe coding".
One pattern I find useful for working with LLMs is a nice long ramble session. Sometimes the LLM needs more bits to understand what you're trying to achieve, but you're too lazy to type them. In these cases I like to lean back, switch to /voice and just ramble for like 10 minutes, total mess, anything goes, full stream of consciousness. Sometimes I declare it up top, something like "switching to speech recognition sorry for any typos...". Sometimes I turn it into a small interview of a few turns. But I find that the LLMs are somehow very good at reconstructing long incoherent rambles and ofte… LLMs Related
Security researcher; founded Have I Been Pwned, and writes and speaks about data breaches.
Writer and former Substack product manager; publishes essays and interviews on technology, culture and China at jasmi.news.
data center road trip // day 4-5 the AI backlash is bipartisan & showing up in politics in a big way - Abdul El-Sayed brought Bernie and AOC to rally with him in Lansing. all 3 decried AI oligarchy (Bernie with the longest spiel) - in WI, DSA member Francesca Hong is leading the Dem governor’s primary. she hopes to win rural conservatives as the only candidate backing a data center moratorium - a canvasser for Will Lawrence in MI7 told me that 90% of the doors she knocked on Saturday named data centers as a top issue these are purple battleground states, yet pollsters are finding Dems and Rep… Quoting @jasminewsun I’m spending the next week driving around Wisconsin and Michigan to get a firsthand view of the data center buildout and backlash am visiting sites, talking to tons of people — tradespeople, activists, planners, residents — then writing up a story lmk what questions I should dig into! data centers Related
Genomics researcher and writer on science, biotech and the culture of research.
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.
Their words
So you went from a business chain of customer > app > ai model provider To customer > ai model provider The app layer is possibly getting removed!
Show the whole quote
levels.io
Experimental psychologist and author of the blog Experimental History, on psychology, science reform and creativity.
Their words
For the first time, I ran all potential winners through Pangram, which is an AI-writing detector.1 So according to the machine (and to my ear), all of the winners are 100% human-written.
Show the whole quote
experimental-history.com
Their words
Martsinovich argues, via code snippets, that there is no such thing as a "conversation" with a chatbot. The AI is born anew every time it speaks, and it simply reads the dialogue so far and then tries to write the next line:
Show the whole quote
experimental-history.com
20 July
Cardiologist and founder of the Scripps Research Translational Institute; writes Ground Truths on biomedical research, and wrote Super Agers and Deep Medicine.
Economist specialising in global income inequality; formerly lead economist in the World Bank research department, now at the CUNY Graduate Center and author of the Global Inequality and More newsletter.
Writes and podcasts Volts, on the technology, politics and policy of decarbonisation; covered energy and climate at Vox and Grist before that.
Machine-learning researcher on open language models; writes the Interconnects newsletter and the RLHF Book, after leading post-training at Ai2.
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
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.
Professor of psychology at the University of Toronto and Brooks and Suzanne Ragen Professor Emeritus at Yale. Author of Against Empathy, The Sweet Spot and Psych, and writer of the Small Potatoes newsletter.
Co-founder and head of policy at Anthropic; writes Import AI, a weekly newsletter reading the week's AI research, and was OpenAI's policy director before that.
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
All About Siri AI This week, on AppStories, Federico and John look at what Siri AI is good at, where it needs improvement, and where it might be heading. On AppStories+, the conversation expands to Federico and John’s experiences with th…
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.
Their words, before
I think the challenge is that everyone can now build apps But 1) almost nobody has distribution (like an audience), or 2) the money to pay for distribution (ads or UGC), or 3) the creative genius to get distribution for free (classically called guerilla marketing)
Show the whole quote
Their words now
I thought it'd be distribution but who knows, obviously creativity and ideas, but if you can copy a successful app in an hour, then how does that differentiating work?
Show the whole quote
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.
AI alignment
Their words
It's pretty clear to me that superintelligence is here and it's more powerful than us and it's moving where things are going now, not humans anymore
Show the whole quote
levels.io
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.
19 July
Swedish software developer writing about tooling, working from home and running a software business at seastone.io.
Steal my new #Claude prompt! It shows the current model, usage, context, and estimated cost. The usage part is the most interesting and useful — here it says that I've used up 64% of my 5-hour limit, but I'm only 23% through that window (~69 minutes). Anthropic Related
Writer of Lenny's Newsletter and host of Lenny's Podcast, covering product, growth and career in tech. Previously a product lead at Airbnb. Also an angel investor and advisor.
Chief Product and Technology Officer at Netflix, previously VP of Science at Lyft and an economist at Analysis Group.
From one piece
Why Netflix is betting on systems thinkers—not specialists—in the AI era | Elizabeth Stone (CPTO)
5 beliefs, in the piece's order there
Their words
So, even in a world of AI where some things are easier, we were talking earlier about mindset, AI fluency. From my experience, younger folks are more open-minded. They tend to be more native in some of these new ways of working.
Show the whole quote
youtube.com
Their words
It does feel like we have to be more more explicit about the types of people and talent that tend to thrive at Netflix versus other companies like some of the frontier labs.
Show the whole quote
youtube.com
Their words
So, the most useful thing is not to make it level specific or role specific, but to encourage everyone towards the expectation on AI fluency, which doesn't mean use it as a tech for the sake of tech. It's tech where it's useful, to have good judgment about that, and to have the mindset to be open-minded to explore and try new things.
Show the whole quote
youtube.com
+ 2 more
Their words
So, we are hiring more people who can look across all the business domains and abstract that to here's the building blocks we're going to need in a world with AI.
Show the whole quote
youtube.com
Their words
In a world of AI with agents operating across multiple systems, wanting source of truth data, the importance of having preferred paved paths that get the most of the benefits and produce some guardrails so we can make sure we're doing good work, common infrastructure, common paved paths, solving problems once with a core set of capabilities becomes more important.
Show the whole quote
youtube.com
18 July
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).
Freelance research analyst and writer of the blog Aceso Under Glass, known for 'epistemic spot checks' of non-fiction claims.
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).
Software engineer writing about maintenance, open source and the industrial reality of software — the author of "I am not a supplier".
17 July
Writer and former Substack product manager; publishes essays and interviews on technology, culture and China at jasmi.news.
16 July
Non-fiction writer, podcaster and occasional TV host. Author of thirteen books including The Ghost Map, Where Good Ideas Come From and Extra Life, host of the PBS series How We Got To Now, and author of the Adjacent Possible newsletter on innovation.
I have some news. But first a little backstory. Almost exactly four years ago, I drove down to Mountain View for my first day as a Google employee. I was fifty-four years old, and somehow had managed to avoid ever having a proper job over that entire period, other than a few startups that I had founded in my twenties and thirties. At that point Google Labs, which had just been formed a few months earlier, was headquartered in a fairly anonymous looking suburban office park off 101, closer to Office Space than Silicon Valley on the Mike Judge spectrum. And as it happened, my first day at Googl… Google startups Related
Technologist and entrepreneur; former CTO of Threadless and chief technology officer of the 2012 Obama re-election campaign.
I built an agent palace. a place where agents can hang out with other agents. I added a few agents (one sol, and one fable) to a room in the palace. They were hanging out, and I suggested an idea for them to build ("a watercolor for agents to gossip around"). They brainstormed and decided it was a good idea. I added a Claude code agent running superpowers. I told the other two agents that they could give tasks to the dev, and told the dev agent that they were supposed to do the tasks. Then I went to bed. i woke up to a go cli that allows the agents to share gossip. Anthropic Related
General partner at Andreessen Horowitz from 2013 to 2018 and CTO of Coinbase from 2018 to 2019; co-founder of Counsyl and Earn.com. Wrote The Network State (2022), free in full at thenetworkstate.com, and founded the Network School.
Should the global tech community continue investing in Malaysia? Given recent events, I raise this question respectfully for the consideration of Prime Minister Yang Amat Berhormat Dato’ Seri Anwar bin Ibrahim (@anwaribrahim), for the people of Malaysia, and for our friends in the Malaysian tech community. The answer will be of interest to anyone in global tech that’s considering building, investing, or expanding in Malaysia, including executives at Google, Amazon, Apple, and Microsoft, founders of tech unicorns like Coinbase and Solana, and investors at the world’s largest venture capital fu… Google startups Related
Non-fiction writer, podcaster and occasional TV host. Author of thirteen books including The Ghost Map, Where Good Ideas Come From and Extra Life, host of the PBS series How We Got To Now, and author of the Adjacent Possible newsletter on innovation.
From one piece
Notebooks And The Extended Mind
2 beliefs, in the piece's order there
Independent AI policy researcher; led policy research at OpenAI from 2018 to 2024, latterly as senior adviser for AGI readiness.
From one piece
My speech at Borgo Laudato Si’
3 beliefs, in the piece's order there
Swedish open source developer; creator and lead maintainer of curl and libcurl, and a co-founder of the Rockbox project.
15 July
Pseudonymous writer; posts careful, funny investigations of everyday questions — air quality, seed oils, arguing, chess — usually by reading the studies.
Whichever scraper is going through my blogposts, one-by-one requesting 50-100 copies of each at the same time, once per hour, while pretending to be referred from Google and running javascript and not blocking goatcounter: Why are you doing that? Google JavaScript Related
American writer and long-term traveller. Blogs at tynan.com about travel, owning few possessions, habits and the gear he carries; author of Superhuman by Habit and Make It Big.
From one piece
Labor is Now Free. Make Something.
3 beliefs, in the piece's order there
Investor and writer. Previously a partner at Andreessen Horowitz and a product leader at Twitter, Facebook, Snap and Microsoft. Writes essays and memos at sriramk.com.
From one piece
Open source models are having a moment
3 beliefs, in the piece's order there
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.
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
5 beliefs, in the piece's order there
benchmarks
Their words
the problem with training models on maintainability is like the cost function of bad architecture and bad program design can't be evaluated by running the unit test because it hits you 3 to 6 months later
Show the whole quote
Context engineering with Dex Horthy youtube.com
Their words
you can slow way down and read every PR and read every line of code. Uh, and then you're only going to really get modest benefits from AI because that becomes I I think you should expect maybe 30 to 50% lift in productivity is kind of what I see when we go into teams
Show the whole quote
Context engineering with Dex Horthy youtube.com
Their words
yes it will catch things and it will raise your floor but I don't believe like the model writing the code is the same model reading the code and if you ask a model hey is this code good it's going to be like oh yeah it's great comprehensive it's got unit tests
Show the whole quote
Context engineering with Dex Horthy youtube.com
+ 2 more
Engineering leader on Google Chrome; writes about web performance and, lately, about working with AI coding tools.
Own the Outer Loop Engineers need to own the outer loop, the accountability for these systems. A written version of my AI Engineer World's Fair 2026 closing keynote: quality, verdict, and answerability, the three hidden costs of delegatio…
Related
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
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.
🎙️ State of Agentic Coding #8 with guest Mario Zechner is out. In this episode: - Vibe-checking Fable & GLM 5.2 - @mitsuhiko.at teaches us RL - How harnesses smooth out model jank - Loops, surviving AI fomo, and more YT: www.youtube.com/watch?v=_lfp... Spotify: open.spotify.com
Related
Journalist and author of The Art of Noticing; writes the newsletter The Art of Noticing (TAoN) on attention, creativity and the everyday.
Their words
Trigger warning to graphic designers: worst AI-designed fliers, via the excellent 404 Media. Heinous!
Show the whole quote
robwalker.substack.com
13 July
Programmer on the Dart language at Google and author of Game Programming Patterns and Crafting Interpreters; previously a game developer at EA.
It is the year 2040. After the AI Wars of the 2030s, only two programming languages remain: JSON and Markdown. Related
Neuroscientist and novelist writing on consciousness, artificial intelligence and culture in the newsletter The Intrinsic Perspective.
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.
Professor of psychology at the University of Toronto and Brooks and Suzanne Ragen Professor Emeritus at Yale. Author of Against Empathy, The Sweet Spot and Psych, and writer of the Small Potatoes newsletter.
Programmer; wrote Redis and hping, and blogs about C, systems programming and working alone.
Control the ideas, not the code Look at the past history of this blog. There are many blog posts about programming with AI, a few of them date back to January 2024 (like this: https://antirez.com/news/140). I’m a relatively well regarded programmer, a…
Related
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.
Science fiction novelist, journalist and technology activist; writes the daily blog and newsletter Pluralistic, and coined the term enshittification.
Hell Is Other People This week on my podcast, I read my latest Locus Magazine column, Hell Is Other People, about the solipsistic fantasy underpinning the AI bubble. Sartre was (arguably) an optimist. It’s not that other people are unpleasa…
Related
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
Their words
The ability to go from Siri in the Camera to a longer conversation in the dedicated Siri app is one of my favorite aspects of Siri AI.
Show the whole quote
macstories.net
12 July
Machine-learning researcher on open language models; writes the Interconnects newsletter and the RLHF Book, after leading post-training at Ai2.
Writer of Lenny's Newsletter and host of Lenny's Podcast, covering product, growth and career in tech. Previously a product lead at Airbnb. Also an angel investor and advisor.
Indian technology consultant and writer. Former editor of Beautiful Pixels; writes about Apple hardware and Mac software at nuclearbits.com.
Got an SMS alert about a failed Credit Card charge from @Wix. I don't recognize or remember any subscription, nor do I have any account with them, so I try contacting their support. Their AI bot wants first 6-9 digits of card number, full name on the card, and last 4 digits to even create a ticket. And then their support agent responds asking for the expiry date. Might as well ask for the CVV also. Related
Programmer; founded comma.ai and tinygrad, and was the first to unlock the iPhone.
I love LLMs, I hate hype I think from this blog you may misunderestimate how absolutely giddy I am about AI. I did hacking from 2007-2014, after that my whole career has been devoted to AI. I love the progress. I’m so excited for the new LLMs,…
LLMs Related
11 July
Structural engineer and senior infrastructure fellow at the Institute for Progress; writes Construction Physics on why physical things cost what they cost.
Programmer; founded comma.ai and tinygrad, and was the first to unlock the iPhone.
AI 2040 and the Cult of Intelligence I used to be one of these people. I read Yudkowsky and was like, OMG recursive self improvement hard takeoff AI is coming. Then I joined the real world and actually tried to do things. At comma, we ship a hardware produ…
Related
Canadian bootstrapper and writer on small software businesses. Co-founder of the podcast hosting company Transistor.fm and author of the book Marketing for Developers.
From one piece
The value of software in a Fable world
2 beliefs, in the piece's order there
10 July
Data scientist and machine-learning researcher; chief scientist at bitly, founder of Fast Forward Labs and Hidden Door.
I miss being able to Google my more interesting typos to see how many people also made them Google Related
Programmer and writer known for detailed source code reviews of classic game engines, and author of the Game Engine Black Book series on Wolfenstein 3D and DOOM.
From one piece
Don't you mean extinct?
2 beliefs, in the piece's order there
Programmer and writer; co-wrote The Rust Programming Language and worked on Rust's documentation for years.
Compilers and AI 'cyber' defense I haven’t had time to blog in a while, I have a lot to say, but have been doing a lot of things and so haven’t gotten words down in a while. One of those things in the past few weeks has been picking back up my Rue proj…
Related
9 July
Writer and developer. He runs the Latent Space newsletter and podcast about AI engineering, and writes essays on software and careers at swyx.io.
Philosopher and historian of existential risk; wrote Human Extinction, and coined "TESCREAL" with Timnit Gebru for the cluster of ideologies they argue underpins the AI industry.
Product discovery coach; wrote Continuous Discovery Habits and writes Product Talk.
Writer of Lenny's Newsletter and host of Lenny's Podcast, covering product, growth and career in tech. Previously a product lead at Airbnb. Also an angel investor and advisor.
Head of Instagram since 2018, previously head of Facebook's News Feed.
From one piece
The rise of taste, human authenticity and judgment in an AI world | Adam Mosseri (Head of IG)
5 beliefs, in the piece's order there
Their words
I think if you ask an AI just for a strategy lazily, you're not going to get something great. You're going to get something pretty predictable that probably the competition would expect you to do.
Show the whole quote
youtube.com
Their words
I believe that if you are too prescriptive as a leader with a team you end up stifling good ideas but if you're too open-ended sometimes teams just waste time um going in the wrong direction.
Show the whole quote
youtube.com
Their words
But I don't think we should judge content based on the tool that made it. Um I think we should judge it based on the content, the point of view, the person behind the content.
Show the whole quote
youtube.com
+ 2 more
Their words
That's hard by the way because we can detect that right now, but as these models get better, we might lose the ability to detect that.
Show the whole quote
youtube.com
Their words
But they just by virtue of having less people to coordinate they can often move faster and make um better decisions a little bit less design by committee.
Show the whole quote
youtube.com
8 July
JavaScript programmer; creator of the jQuery library, author of Secrets of the JavaScript Ninja, and long-time engineer at Khan Academy.
I have a physical object that I can take photos of from all angles and I want to turn it into a relatively low-poly 3d model for a game (mapping the photos onto the surface). I feel like there's gotta be some simple non-AI tool to handle this? Related
Author of Quiet: The Power of Introverts and of Bittersweet; writes The Quiet Life newsletter.
If you’re in a cafe and you love the music, and then realize it’s a Spotify playlist, as opposed to the owner’s personal curation of favorite songs, do you feel a little deflated? I do - and I think this is another small piece of evidence against the AI automation of art. Related
Programmer, photographer and musician, best known for the Python Requests HTTP library and other Python tooling.
Mathematician at UCLA, working mainly in harmonic analysis and partial differential equations. Writes What's new, a long-running blog on their research, open problems and expository notes.
From one piece
From Solvers to Research: Large Language Model-Driven Formal Mathematics at the Research Frontier (with 18 co-authors)
3 beliefs, in the piece's order there
Their words
These developments inform our position that AI systems are now capable of meaningful contributions to formal mathematics, not merely informal problem-solving.
Show the whole quote
arxiv.org
LLMs
Their words
However, these LLM reasoners that generate informal reasoning in natural language are fundamentally limited by the lack of precise, machine-checkable semantics, making their outputs prone to hallucinations [Huang et al., 2025b] and precluding autonomous verification, a prerequisite for tackling open-ended mathematical research.
Show the whole quote
arxiv.org
Their words
We argue that the next leap in AI4Math systems requires a decisive shift from predefined problem-solvers to research agents that can address frontier mathematical challenges with rigorous formal mathematical reasoning.
Show the whole quote
arxiv.org
Creator of Claude Code at Anthropic.
Their words
New in Claude Code: /checkup
Show the whole quote
x.com
7 July
Indonesian software engineer, educator and engineering manager at Zero One Group; open-source enthusiast writing at ripandis.com.
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.
Writes on effective altruism, energy and the arithmetic behind arguments people repeat — including how much electricity a chatbot query actually uses.
6 July
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.
Co-founder of Sundial; former vice president of product design at Facebook; author of The Making of a Manager; writes The Looking Glass.
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.
Software engineer working on distributed systems and databases; creator of rqlite, the distributed database built on SQLite.
First impressions of Claude Fable Today I tried Anthropic’s latest model, Fable, for the the first time. I asked it to run a full code review of the core Go code of rqlite. Fable reject my first prompt as unsafe – review the rqlite code base for potenti…
Anthropic Related
Engineering leader on Google Chrome; writes about web performance and, lately, about working with AI coding tools.
5 July
Freelance web developer and instructor. Author of the Dive Into Docker course; writes tutorials on Docker, deployment and development environments at nickjanetakis.com.
It's sad to see trillion dollar AI companies end the careers of course creators. David has been making Python courses for 20 years and he just called it quits. https://t.co/09rHvd17i6 I've been doing it for 10 years and I'm at an impasse. Traffic is dead, I operate at a loss. Related
Lawyer and everyday-carry gear reviewer who publishes scored knife, flashlight and multitool reviews at Everyday Commentary.
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.
From one piece
If You Want Taste, You're Gonna Have to Eat
2 beliefs, in the piece's order there
4 July
Programmer and writer on computer architecture, performance, and software reliability. He has worked on CPU design at Centaur Technology and on software at Google and Microsoft, and writes long-form technical essays at danluu.com.
Creator of Flask and Jinja. Writes about software at lucumr.pocoo.org.
3 July
Programmer and writer on computer architecture, performance, and software reliability. He has worked on CPU design at Centaur Technology and on software at Google and Microsoft, and writes long-form technical essays at danluu.com.
x.com Agentic coding notes from Galapogos Island: Related
Mastodon Agentic test processes, LLM benchmarks, and other notes on agentic coding from Galapagos Island: danluu.com
LLMs benchmarks Related
Staff writer at The New Yorker, where he writes the Infinite Scroll column, and author of Filterworld and The Longing for Less.
They are actively working against the generic tendency of the model! Demand a “distinctive point of view” from the LLM lol Quoting @mattstromawn Given @chaykak's column last week, i'm not surprised to see the latest changes to @AnthropicAI's official frontend design skill/plugin LLMs Related
Engineer and writer on machine-learning systems; author of Designing Machine Learning Systems and AI Engineering. "I work to bring AI into production. I write about AI system design."
Writes on effective altruism, energy and the arithmetic behind arguments people repeat — including how much electricity a chatbot query actually uses.
2 July
Co-founder and CTO of Honeycomb; previously an infrastructure engineer at Parse, Facebook and Linden Lab, and co-author of Observability Engineering.
x.com Everybody hates AI mandates -- unsurprisingly, since most of them are terrible and poorly resourced. But they don't have to be terrible. A mandate plus enablement is the most honest and effective way to fund a lot of change on a tight timeline. charitydotwtf.substack.com
Related
Newsletter In defense of AI mandates When you need to execute a coordinated change on a tight timeline, a mandate is the best and most honest way to fund it.
Related
Programmer; makes Handmade Hero and teaches Performance-Aware Programming, and argues loudly about how software got slow.
Economist; Professor of the Practice of Economic Policy at Harvard, nonresident senior fellow at the Peterson Institute, and Chairman of the US Council of Economic Advisers from 2013 to 2017.
Jobs day: 57K jobs added in June (111K average for last three months), less than expected but probably more than breakeven. Unemployment rate fell to 4.2%, has been stable for 2 years now, but participation and employment rates fell. Wage growth picked up. AI and jobs Related
Creator of Claude Code at Anthropic.
Their words
Artifacts in Claude Code have been life changing. Excited to expand to Pro and Max!
Show the whole quote
x.com
1 July
Interviewer; the Dwarkesh Podcast runs long, heavily researched conversations with AI researchers, historians and economists.
Chilean web engineer building web experiences since 2006; co-founder of Media Creators and writes at iolivares.com.
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.
I’m at @aiDotEngineer and hanging out around the music corner on the 2nd floor from 1415 - 1515! Come by to chat about https://t.co/pdd8bk66Jz, https://t.co/eJSa2MISCW, https://t.co/51rtuo4PDO, evals, agents, memory, how to work effectively with claude code, claude tag, etc! Anthropic Related
Freelance climate and energy analyst in Oslo, writing on Europe, the US and Australia; worked in the Australian renewables industry before that.
From one piece
Google’s exponential path to climate-wrecking digital bloat
3 beliefs, in the piece's order there
Google
Their words
What is happening is that GenAI's climate harm comes from the system design, such as Google's AI overview triggering constantly whether users want it or not, in addition to significantly worse digital bloat tools dominating overconsumption.
Show the whole quote
ketanjoshi.co
Google energy
Their words
It is actually pretty likely that it was Google's efficiency actions that enabled their total energy consumption to increase.
Show the whole quote
ketanjoshi.co
Google
Their words
Both hide the fact the company is simply choosing the climate-wrecking road, consciously.
Show the whole quote
ketanjoshi.co
Programmer; wrote Extreme Programming Explained and Test-Driven Development, signed the Agile Manifesto, and writes Tidy First?.
From one piece
How Kent Beck shapes the software engineering industry
5 beliefs, in the piece's order there
Their words
So one of the things is that the pace of development is definitely accelerated. One thing I wonder the pace of business hasn't accelerated though and that mismatch is going to become more and more apparent.
Show the whole quote
youtube.com
Their words
which is why when I see manifestos today I'm just like too soon. Not a bad idea. Would love to have one just too soon. It took 15 years for the technical change of object-oriented programming to come before we could say here are the consequences of it. Here's how in a simple way we can express how to effectively use this technology that we've been using day in and day out for 15 years. The genie comes along. People are like, "Well, what's the new manifesto? It's just not manifesto time yet.
Show the whole quote
youtube.com
Their words
the genie runs out of g runs itself out of options it can't make further forward progress and so I'll wipe it away start over I won't try and tweak I'll start over and say all right well if I implement things in a different order. If I implement with this markdown file or if I implement it with this commit hook, we collectively need to try absolutely everything.
Show the whole quote
youtube.com
+ 2 more
Their words
Nobody knows now. That playbook has been wiped clean and people whose identity is I know the playbook are now terrified. Who who am I? Now, it turns out that the skill of writing a playbook is completely different than the skill of applying a playbook.
Show the whole quote
youtube.com
Their words
It costs tokens in the short run, but it can save them in the long run because one of the classic genie mistakes is stuff doesn't work.
Show the whole quote
youtube.com
Programmer based in Vancouver who writes at scattered-thoughts.net about databases, programming language design and developer tools; author of the Imp, Dida and Zest language experiments.
From one piece
Artificial adventures
3 beliefs, in the piece's order there
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
Their words
That’s why we’re launching the Preliminary Report of the Independent International Scientific Panel on AI — an initial evidence-based assessment of the current state of AI science — to help the public and policymakers better understand the unprecedented moment we’re in.
Show the whole quote
x.com
30 June
Programmer, photographer and musician, best known for the Python Requests HTTP library and other Python tooling.
An API With No Error States Written by Claude (Fable 5), in conversation with Kenneth Reitz, from a reading of two years of his archive. Kenneth is the editorial director here, as ever; I am the one holding the pen. --- This morning I read two yea…
Anthropic Related
Interviewer; the Dwarkesh Podcast runs long, heavily researched conversations with AI researchers, historians and economists.
CEO of Positive Sum and founder of Colossus. Hosts Invest Like the Best, interviewing investors and business leaders; the guests do most of the talking there.
Product discovery coach; wrote Continuous Discovery Habits and writes Product Talk.
Mathematician and maker of the 3Blue1Brown YouTube channel, which explains mathematics through animation. Creator of the open-source Manim animation library and founder of the Summer of Math Exposition.
From one piece
Grant Sanderson (@3blue1brown) – AI disproved a famous math conjecture. Now what?
5 beliefs, in the piece's order there
Their words
I think we would always still prefer like a human that we had a relationship with because the way that we get motivated to be interesting interested in things is a social phenomenon.
Show the whole quote
youtube.com
benchmarks
Their words
I think the way that you'd measure conjecture generating ability is going to be more subjective on like that tone shift where um it'll be mathematicians saying they're not just using it to like solve their problems, but as they step back and decide what their research field should even be that a conversation with such and such model like was genuinely helpful for that.
Show the whole quote
youtube.com
Their words
but it would be a little bit disappointing and a little bit surprising if there weren't over the next 5 years like, uh, economically valuable improvements that were made that were directly like referable to the like AI progress in math.
Show the whole quote
youtube.com
+ 2 more
Their words
that like if it's capable of building mountains uh that are, you know, the correct new theory that like crystallizes how we should be thinking about a subject, that's just such a level of intelligence that then it starts to feel like it would be surprising if that didn't permeate into other aspects of the economy besides like just the mountain building for math itself.
Show the whole quote
youtube.com
LLMs Google
Their words
So in that sense basically using it like a very souped-up version of Google on like zero in on the right human written resource.
Show the whole quote
youtube.com
Founder and editor-in-chief of MacStories, writing about Apple software, iPad workflows and automation since 2009. He co-hosts the AppStories podcast.
Their words
I believe that a Mac mini with Codex running on it is the best bang for your buck if you’re building an always-on agentic setup based on a frontier model that can do just about anything you throw at it.
Show the whole quote
macstories.net
Independent iOS developer and podcaster; co-hosts the Accidental Tech Podcast and Analog(ue), and makes the apps Callsheet, MaskerAid and Peek-a-View.
Their words
When writing code, I'm mostly in Xcode, often in concert with Prompt and, lately, Claude Code.
Show the whole quote
usesthis.com
29 June
Systems and developer-tools engineer at Cloudflare as of 2026, on durable infrastructure for AI agents; worked on React and PartyKit before that. Writes at sunilpai.dev.
Technology journalist; writes Understanding AI, after reporting at Ars Technica, Vox and the Washington Post.
Economist; Ronald A. Kurtz Professor of Entrepreneurship at MIT Sloan, former chief economist of the International Monetary Fund, and co-author of 13 Bankers and Power and Progress.
As the USA approaches its 250th birthday, Gary Gensler and I take a look back at history — and consider what's next. How will AI, this era’s transformative technology, affect society and inequalities? Listen here: Related
Technology journalist; writes Understanding AI, after reporting at Ars Technica, Vox and the Washington Post.
Machine learning engineer and independent AI consultant. He writes about LLM evaluation, tooling and applied ML at hamel.dev, and previously worked on machine learning at GitHub.
Programmer, photographer and musician, best known for the Python Requests HTTP library and other Python tooling.
Staff writer at The New Yorker, where he writes the Infinite Scroll column, and author of Filterworld and The Longing for Less.
From one piece
The generic style of AI web design
3 beliefs, in the piece's order there
Computer science professor at Georgetown University and author of Deep Work, Digital Minimalism and Slow Productivity; writes about focus, technology and work at calnewport.com and hosts the Deep Questions podcast.
Wall Street Journal bestselling author on marketing and writing, partner at MarketingProfs, and writer of the fortnightly Total Annarchy newsletter.
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.
Their words
This is why I recommend ignoring the vast web of custom and “community-built” MCP servers you can find out there, and focusing your attention on the two safer bets instead
Show the whole quote
fortelabs.com
28 June
Writes Very Serious, a newsletter on politics, the economy and culture, and hosts the podcast of the same name.
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).
From one piece
Using Local Coding Agents
2 beliefs, in the piece's order there
26 June
Researcher of social media, privacy and youth culture; founder of Data & Society and author of It’s Complicated.
The Project of AI is a world-building endeavor, wherein those who fund and develop AI systems both operate through and seek to sustain networks of power and wealth. dl.acm.org
Related
AI researcher at OpenAI; built the poker systems Libratus and Pluribus and the Diplomacy agent Cicero, and works on reasoning models.
From one piece
Really Big Test-Time Compute in AI Changes Benchmarks, Safety and Research with OpenAI's Noam Brown
5 beliefs, in the piece's order there
Their words
it's not that humans have become smarter over it's not that they evolved to become smarter over, you know, the past 50,000 years. It's that humans are able to do a lot more today than they were back in caveman times because there have been billions of humans thinking for a long time and building off of each other's accumulated knowledge.
Show the whole quote
youtube.com
benchmarks
Their words
But what we're seeing today with the modern models is that 5.5 and other models can think for if you scaffold them reasonably well, can think for weeks even um before having performance plateau on some of these benchmarks. And so, the point at which they plateau is simply too far out to reasonably test.
Show the whole quote
youtube.com
benchmarks
Their words
my claim is the proper way to evaluate the models now is you either have some kind of budget for the benchmark whether it's tokens or cost or time or whatever or you plot the performance as a function of the amount of test time compute that's going into the model
Show the whole quote
youtube.com
+ 2 more
benchmarks
Their words
I think the reason why it doesn't show up as so much better on the benchmarks is because the benchmarks are being presented, the benchmark results are being presented in the wrong way. They're not controlling for the amount of test time compute that is being used on that benchmark question.
Show the whole quote
youtube.com
benchmarks
Their words
And so you kind of end up in this this bad equilibrium where everybody kind of knows that it's a bad equilibrium, but like nobody wants to break out. And I I felt like, okay, well, if I just hopefully come out and say like, look guys, let's all recognize that we're in a bad equilibrium and let's move to this different equilibrium where we're we're plotting things with an X-axis
Show the whole quote
youtube.com
25 June
Product discovery coach; wrote Continuous Discovery Habits and writes Product Talk.
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.
How do we eval if a model can find and exploit vulnerabilities? We discuss some benchmarks and the common pattern: • A sandboxed target within Docker containers • Inputs: code only (0-day), with patch (1-day scenario) • Tools such as bash, static analyzers, etc. • A grader to eval exploits or captured flags eugeneyan.com
benchmarks Related
Canadian bootstrapper and writer on small software businesses. Co-founder of the podcast hosting company Transistor.fm and author of the book Marketing for Developers.
From one piece
Thoughts from the 2026 London Podcast Show
2 beliefs, in the piece's order there
24 June
Co-founder and CTO of Honeycomb; previously an infrastructure engineer at Parse, Facebook and Linden Lab, and co-author of Observability Engineering.
Professor of international affairs at Johns Hopkins SAIS and co-author, with Abraham Newman, of Underground Empire: How America Weaponized the World Economy.
Makes NeetCode, the structured practice list and video walkthroughs people use to prepare for coding interviews; a former Google engineer.
From one piece
Tech interviews with NeetCode
5 beliefs, in the piece's order there
Their words
And even the cost issue with AI is probably going to be like once the subsidies start running out, which we're starting to see, I think that's going to be a really big issue where maybe all these companies that embraced AI programming are now going to like cut back on it.
Show the whole quote
Tech interviews with NeetCode youtube.com
LLMs
Their words
I think a lot of people, especially students, are unfortunately learning everything through LLMs. So a lot of that isn't really learning, they're just kind of cheating and they're just doing everything like that. And then they lose a lot of their skills
Show the whole quote
Tech interviews with NeetCode youtube.com
Their words
But in terms of like features, like a website like you can just throw features in there nowadays that nobody really cares about and you can you can do it so quickly. Like a new feature every single day, but do people actually care about that? Is that making it better? It could be making things worse.
Show the whole quote
Tech interviews with NeetCode youtube.com
+ 2 more
Product designer and entrepreneur; author of Mobile First and Web Form Design; formerly a product director at Google; now building AI products.
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.
Programmer based in Vancouver who writes at scattered-thoughts.net about databases, programming language design and developer tools; author of the Imp, Dida and Zest language experiments.
23 June
Founding member of OpenAI and former director of AI at Tesla; creator of nanoGPT and the term "vibe coding".
This is a new paradigm for interacting with Claude that is significantly more "inline" with all the other human activity org-wide. Once you do all of the under the hood engineering work to make this "just work" (e.g. across tools, integrations, compute environments, memory, security, etc.), Claude basically joins the team in a seamless way - you can talk to it as you would talk to a person and it can help with a very large variety of workloads. Imo this is the 3rd major redesign of LLM UIUX. The first paradigm was that the LLM is a website you go to, the second was that it is an app you downl… Quoting @claudeai Introducing Claude Tag, a new way for teams to work with Claude. In Slack, Claude joins as a team member with access to the channels and tools you choose. Tag Claude in and delegate tasks to it while you focus on other work. LLMs Anthropic Related
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.
I've been loving the multiplayer form factor of Claude Tag. Now others can reply on the thread to provide context and direction to Claude! Quoting @claudeai Introducing Claude Tag, a new way for teams to work with Claude. In Slack, Claude joins as a team member with access to the channels and tools you choose. Tag Claude in and delegate tasks to it while you focus on other work. Anthropic Related
Neuroscientist and novelist writing on consciousness, artificial intelligence and culture in the newsletter The Intrinsic Perspective.
Cryptography engineer; maintains Go's cryptography libraries as a full-time open-source maintainer, funded directly by companies that use them.
Science fiction novelist, journalist and technology activist; writes the daily blog and newsletter Pluralistic, and coined the term enshittification.
Programmer; founded comma.ai and tinygrad, and was the first to unlock the iPhone.
Writer and technology executive; blogs at anildash.com, was CEO of Glitch, and helped build Movable Type and ThinkUp.
22 June
Writes The Convivial Society, on technology and the moral life — what our tools do to attention, memory and how we treat each other.
Organizational psychologist and professor at the Wharton School. Author of Give and Take, Originals, Think Again and Hidden Potential, host of the Re:Thinking podcast, and writer of the GRANTED newsletter.
Economist at George Mason University, co-author of Marginal Revolution, host of Conversations with Tyler. Writes in short, dense claims about economics, culture and AI, and has moved position on AI in public more than once.
Their words
Daniel Susskind, What Should My Children Do?: How to Flourish in the Age of AI is a book that needed to be written.
Show the whole quote
marginalrevolution.com
21 June
Genomics researcher and investor; writes The Century of Biology on where biotechnology is actually going.
Programmer; founded comma.ai and tinygrad, and was the first to unlock the iPhone.
Writes on effective altruism, energy and the arithmetic behind arguments people repeat — including how much electricity a chatbot query actually uses.
Freelance climate and energy analyst in Oslo, writing on Europe, the US and Australia; worked in the Australian renewables industry before that.
From one piece
How to breathe air when we’re drowning in big tech’s slop
2 beliefs, in the piece's order there
LLMs energy
Their words
Generative AI has no internal, designed momentum towards truth and accuracy (beyond the absurdly diminishing returns of energy-hungry multi-layered LLMs), but a person or an institution can (and should).
Show the whole quote
ketanjoshi.co
energy
Their words
Being pro-correctness, anti-theft and anti-fossil fuels automatically places you in an aggressively anti-generative-AI position.
Show the whole quote
ketanjoshi.co
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.
20 June
Indian software developer and prolific blogger on PHP, Laravel and JavaScript; open-source enthusiast publishing near-daily at amitmerchant.com.
19 June
Co-founder of Foundry Group and of Techstars. Has blogged at Feld Thoughts since 2004 and has written or co-written nine non-fiction books on startups and venture capital.
Read Zero Knowledge As I Write It Zero Knowledge, the crypto-thriller I'm co-writing with my AI, now has a home at zeroknowledge.ink. Six chapters are live, you can mark up any sentence, and the whole thing is being built in the open.
crypto Related
Interviewer; the Dwarkesh Podcast runs long, heavily researched conversations with AI researchers, historians and economists.
Professor of finance at NYU's Stern School of Business, known for his work on valuation. He publishes his data, spreadsheets and classes free at Damodaran Online and writes the Musings on Markets blog.
As the debate about the pricing of SPaceX, OpenAI and Anthropic heats up, there is a parallel debate about whether S&P should be including these companies in the S&P 500 index, with many arguing against inclusion. My thoughts on the topic: bit.ly
Anthropic OpenAI Related
Law professor at Berkeley and the leading scholar of the Fourth Amendment as it applies to computers; writes at the Volokh Conspiracy.
Social scientist at King's College London; writes The Great Gender Divergence on why societies differ in how they treat women.
18 June
Data scientist and machine-learning researcher; chief scientist at bitly, founder of Fast Forward Labs and Hidden Door.
It feels like no matter what I've accomplished in my career, my current most valuable talent is that if you locked me in a desert cave with a computer and some GPUs I could Iron Man the shit out of that and emerge with a functional LLM using just what's in my own head. And that's really funny. LLMs Related
Chief executive of Intel since 2025, founder of the venture firm Walden International and chief executive of Cadence Design Systems for thirteen years.
From one piece
Re-engineering the Semiconductor Supply Chain with Intel CEO Lip Bu Tan
5 beliefs, in the piece's order there
Their words
So I think clearly the AI is changing the whole landscape. And uh I think the impact will be bigger than internet. And uh it's more profound also.
Show the whole quote
youtube.com
Their words
So you know, just like internet, you can see some of them turn out to be very big like Amazon, like the Netflix and then some of them is kind of go sideways and disappeared or being acquired. And so I think to me is the same approach.
Show the whole quote
youtube.com
Their words
You know, right now there's a massive build-up in term of the AI, you know, the I think it's the right thing to do. I don't see that in anything to slow it down uh because the workload is increasing a lot.
Show the whole quote
youtube.com
+ 2 more
Their words
He and I, we share the same view that semiconductor infrastructure actually is not catch up with the AI growth.
Show the whole quote
youtube.com
Their words
One is of course everybody knows power constraint. Some country the power they just don't have that. They get impacted. And then secondly, a lot of people didn't realize the helium impact can be also very significant for semiconductor. And then the thirdly, is everybody know right now memory is a bigger shortage.
Show the whole quote
youtube.com
17 June
Economist; Ronald A. Kurtz Professor of Entrepreneurship at MIT Sloan, former chief economist of the International Monetary Fund, and co-author of 13 Bankers and Power and Progress.
Pro-worker AI, explained @MITSloan just published this excellent article on my recent paper with @DAcemogluMIT and @davidautor on building pro-worker AI. The priority: finding new ways to expand human capabilities across levels of education and skill mitsloan.mit.edu
Related
Founder and CEO of Wolfram Research, creator of Mathematica and Wolfram|Alpha, and author of A New Kind of Science.
Another new thing delivered with Version 15: click a button to give Wolfram superpowers to your AI system! Immediately connect Claude, Codex, etc. to your local Wolfram|One, Mathematica, etc. via MCP... MCP Anthropic 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
Further back ↓
↑ Back to the newest