Related posts

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.

• It’s been 2 hours and I can already tell this is going to be my most popular post. See the replies. And I’m getting so many DMs from people who went through the same, I’m glad it’s connecting. 2 quick things - - please consider the solution!! it’s counterintuitive and no one considers it but it’s super effective - I hope managers see this too and use it as a tool to help their reports
  Quoting@threepointonenew post: the senior engineer death spiral https://t.co/xzgU1jVBRh a friend just started a big job and asked for some advice. so I braindumped a monologue about a super common failure mode I see with engineers and posted it here, hope it helps whoever it can.

  Related

Science fiction novelist, journalist and technology activist; writes the daily blog and newsletter Pluralistic, and coined the term enshittification.

• til the printscreen function in #Ubuntu saves the screenshot to ~/Pictures/Screenshots as well as letting you copy to the clipboard (I just deleted about 8,000 saved screenshots!) (then I rebound printscreen to Flameshot)

  Related

Singaporean writer and author of FRIENDLY AMBITIOUS NERD; blogs at visakanv.com about ambition, writing and thinking in public.

• good news: I’m still available for marketing consults (see pinned tweet)
  Quoting@pmituterrible news: Marketing (even with the use of AI) is still super hard

  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.

• So many people in the replies saying it's overcomplicated, and no proof that this works And then plenty other folks saying that at their startup/company they built this the below pretty much, and then a few new things Different speedlanes when it comes to agents + pipelines
  Quoting@GergelyOroszHere's what OpenAI's agentic software factory looks like, today. Details: https://t.co/UUZufLxG2r (thanks to all the OpenAI folks who explained how it works! And Perf Factory looks especially interesting to me)

  startupsRelated

Geneticist and writer on human population genetics, deep history and evolution; author of the Unsupervised Learning newsletter.

• this sort of tweet is fine. i have the same concerns
  Quoting@BlakeSNeffI think this is one of the most important facts to be aware of in the debate over Indian immigration: Even relative to other Asian immigrants, Indian Americans are on average very liberal, not just in terms of voting Democrat but also holding particularly left-wing views on specific issues. Indian…

  Related

• so what should i do about it? not tweet about genetics? 🤷🏾‍♂️
  Quoting@ProfBootyPhD@razibkhan "As usual when i get surfaced in some parts of X" - you are too generous to your followers (understandably). Almost every time you post about anthropology and genetics, the same people are there to slurp up every factoid that might support white superiority.

  Related

Web standards author and speaker; wrote the CSS: The Definitive Guide series and co-founded the An Event Apart conference.

• I just deleted a post I made yesterday, because it led some people to reasonably read things into it that I didn’t intend. That’s on me, as I wrote it badly and without considering contexts beyond what I was thinking at the time. My apologies to anyone who read it and was disheartened or upset; I’m happy to discuss further in DMs with those who saw it and would like clarification on what I meant.

  Related

Co-creator of Django and creator of Datasette; writes daily at simonwillison.net.

• The contagion of fear

  Bryan Cantrill responds to the tweet by former Anthropic employee Jacob Coxon confirming that many Anthropic researchers believe AI "could kill us all by the end of the decade". Bryan shares a story of his own youthful…

  AnthropicRelated

Economist and journalist; Financial Times columnist as The Undercover Economist, author of How to Make the World Add Up, and presenter of the Cautionary Tales podcast.

• My tweet was less about how to do the task, and more about what happens when I tried to use a well-established tool (Google search). Also - isn't the very high failure rate interesting? 'Sometimes makes errors' is one thing. 'Always makes errors' is something else.
  Quoting@joshgans@TimHarford You need to use a real AI agent to do that task.

  GoogleRelated

Creator of Flask and Jinja. Writes about software at lucumr.pocoo.org.

• Interpreting Pangram

  Original prompt to generate a David Sacks like tweet

  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.

• I’m going to delete this tweet and post screenshots here for posterity. I stand by the argument but we shouldn’t fight. This is too important for squealy arguing. I am going to choose to see David’s OP as an olive branch and operate under that assumption. TYFYATTM!

  Related

Deep-learning researcher and Turing Award laureate; founder of the Mila institute.

• Over the past few days, I've taken the time to summarize my thoughts on the recent incidents involving agents’ misaligned behavior. We don't know with certainty what comes next, but we know where these issues originate, and this can help us plan the path forward. Please feel free to ask your questions in the replies, and I’ll try to answer some of them in the coming weeks.

  yoshuabengio.org

  AI alignmentRelated

Creator of Homebrew, the macOS package manager, and a Swift developer.

• Clicking the Grok button is rarely satisfying. If the tweet is “I like grooglefabs” Grok will delightfully tell you “FooBar comments on his love of grooglefabs. A common sentiment among the fabblegrooglers”. I don't want to type “wtf is a grooglefab”. You are meant to be AGI.

  AGIRelated

Programmer on the Dart language at Google and author of Game Programming Patterns and Crafting Interpreters; previously a game developer at EA.

• RE: https://climatejustice.social/@chu/117242750410634529 I love this toot and all the replies.

  Related

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.

• people reviewing jasmines tweet before the culture fit interview
  Quoting@jasminewsunI hear roughly 3 reasons people keep working at AI labs despite believing in ~10% extinction risk: 1) Techno-determinism: Someone will build ASI no matter what, and I can do it better & more safely than China/OpenAI/etc 2) Consequentialism: ASI might kill us, but it also might produce utopia/immort…

  Related

Writer of the newsletter bookbear express, on psychology, relationships, non-attachment and taste; based in San Francisco.

• just realized replies to my substack posts to have been going to spam for, oh, about a year. SORRY IF YOU EMAILED ME 💔💔💔

  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.

• I use this Vinod Khosla tweet with apologies to Vinod, since tweets are often out-of-context or emotive, to launch a discussion on the trade-off between scaling up and building business models (profitability) that every business faces.

  Related

Economist; former chief economist of the International Monetary Fund.

• another tweet written a bit too fast. By "central bank debt" I mean public debt held by the central bank". Apologies☹️
  Quoting@ojblanchard1No, and useful to repeat: Contrary to interventions such as QE or debt management by Treasury, cancellation of central bank debt has zero mechanical effect on the liabilities of either the central bank (bank reserves are the same) or the government (the debt held by private investors is the same)

  public debtinterest ratesRelated

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.

• Curious why machine learning is not applied to the X timeline more? If somebody is replying once a minute for 8 hours they should be flagged. When I look at reply bot replies its super obvious
  Quoting@simonwQuestion for people who run autoreply bots here: are you not worried about the negative impact they have on your professional reputation? Anyone checking your profile here - a potential future employer for example - will instantly be able to tell you automated replies with a bot

  Related

Developer and podcaster; created Instapaper and Overcast, co-founded Tumblr, and co-hosts the Accidental Tech Podcast.

• RE: https://infosec.exchange/@clayton/117175171619298911 It's OK. This feature is awesome, and one I use heavily. Every social network needs a few mutes here and there. Nobody inherently deserves anyone’s attention. If someone replies in a way that makes me think, "I'd rather not hear from this person again,” I just mute forever. No aggressive block, no report or feedback, no way for them to know. It's not meant to send a message, get revenge, or influence anyone else — just tailor my feed for me.

  Related

American software engineer; worked on code-server at Coder and teaches TypeScript and open-source contribution. Publishes at joeprevite.com.

• today is the day you stop doomscrolling stay with me - this will take 5mins 1. open the AppStore 2. download "ScreenZen" (free) 3. configure limits (e.g. 4 15-min sessions) 4. start with 1 social media app 5. a few days later, add another 6. repeat until you've added all apps too busy right now? add a calendar event with this tweet and do it this weekend trust me, you'll thank me next month

  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?

• i'm always so excited to see that there are new replies to my slack threads until i realize it's just claude (sorry claude)

  AnthropicRelated

Writes jvns.ca, turning "hard and scary" topics — networking, debugging, Linux internals — into plain explanations and hand-drawn comics. Publishes Wizard Zines, illustrated programming zines. Software developer based in Montreal.

• how do you think about how SSH public key authentication works (versus passwords) and what about your mental model makes you trust it? please only answer if you _don't_ understand the math of how RSA (or elliptic curves etc) works, I'm interested in exploring what it means to understand the interfaces around public key authentication without actually learning number theory (also obviously do not try to explain the math to me, also don't explain to folks in replies unless specifically asked)

  Related

Software engineer; co-founder and former CTO of Tailscale, previously on the Go team at Google.

• this old tweet is getting some likes (why? mysterious), so I re-read it, and still like it. it also applies really well to what I am working on now: I need lots of VMs for agents, but I don't need lots of CPU/RAM. hence @exedev
  Quoting@davidcrawshaw@danluu one of the big early ideas behind tailscale was a single computer is powerful enough to solve almost every problem, and that this is not widely understood. we need fewer distributed systems and more spreadsheet implementations.

  Related

Brazilian software engineer, creator and maintainer of GoReleaser; works on Copilot CLI at GitHub. Writes about tooling and his setup at carlosbecker.com.

• Underrated tweet I think it also applies to any software, so long it has users
  Quoting@solomonstreRule #1 of open-source: no is temporary, yes is forever.

  Related

Designer and engineer; formerly principal research engineer at GitHub Next; writes at wattenberger.com.

• there's something rotten in the air when >90% of the replies to a tweet like this are mean spirited sure, it might not be your thing and it's easy to forget this box doesn't go straight into the void. but can we try to bring more kindness into the world?
  Quoting@WattenbergerI look away for one week and VSCode got cute?!

  Related

Indian technology consultant and writer. Former editor of Beautiful Pixels; writes about Apple hardware and Mac software at nuclearbits.com.

• I saw Atul's tweet and it made me wonder too. How many establishments are actually compliant? So we asked the FDA. Here's what they say!

  mid-day.com

  Quoting@atulkasbekarSerious question Of all the raids that were carried out on food establishments, were there any places that were given a clear green chit? As in, everything is perfect, well done! I’d really like to know which places were adhering to standards

  Related

Writer of the personal-finance blog Of Dollars And Data and author of Just Keep Buying and The Wealth Ladder. Chief Operating Officer at Ritholtz Wealth Management.

• The Problem with Optionality

  I’ve seen a lot of discussions this year around maximizing freedom and optionality. For example, consider this tweet: If you ever get up $1M in cash Buy a new modern apartment in the UAE for like $550k (you have zero pr…

  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.

• prompt-cardio — Prompt Faster: the vibe coding interview. A typing test disguised as an AI chat.

  Commitsfeat(ui): tweet-sized results card · feat: harder rank curve — CTO of Vibes now demands 140+ wpm · feat: swarms are engine-injected at rounds 2/4/6 — skill unlocks the spectacle

  Related

Swedish essayist who writes Escaping Flatland, a newsletter of essays on relationships, thinking, writing and homeschooling.

• How not to forget what matters

  1. Every few months I will read a tweet, or have a conversation, that makes me feel this is important, I must remember this.

  Related

Co-founder of AngelList. Writes and podcasts at nav.al in short, aphoristic pieces on startups, wealth and judgement rather than long essays.

• A Return to Code

  You’re listening to the Naval Podcast. This is Nivi, his regular co-host. Today we’re going to talk about vibe coding. Let me tee up the conversation with a tweet from Naval from March 23rd: “AI coding agents can now de…

  coding agentsRelated

Founding member of OpenAI and former director of AI at Tesla; creator of nanoGPT and the term "vibe coding".

• Farzapedia, personal wikipedia of Farza, good example following my Wiki LLM tweet. I really like this approach to personalization in a number of ways, compared to "status quo" of an AI that allegedly gets better the more you use it or something: 1. Explicit. The memory artifact is explicit and navigable (the wiki), you can see exactly what the AI does and does not know and you can inspect and manage this artifact, even if you don't do the direct text writing (the LLM does). The knowledge of you is not implicit and unknown, it's explicit and viewable. 2. Yours. Your data is yours, on your loca…
  Quoting@FarzaTVThis is Farzapedia. I had an LLM take 2,500 entries from my diary, Apple Notes, and some iMessage convos to create a personal Wikipedia for me. It made 400 detailed articles for my friends, my startups, research areas, and even my favorite animes and their impact on me complete with backlinks. But,…

  LLMsRelated

• Wow, this tweet went very viral! I wanted share a possibly slightly improved version of the tweet in an "idea file". The idea of the idea file is that in this era of LLM agents, there is less of a point/need of sharing the specific code/app, you just share the idea, then the other person's agent customizes & builds it for your specific needs. So here's the idea in a gist format: https://t.co/NlAfEJjtJV You can give this to your agent and it can build you your own LLM wiki and guide you on how to use it etc. It's intentionally kept a little bit abstract/vague because there are so many directio…
  Quoting@karpathyLLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as mar…

  LLMsRelated

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.

• It was an honor to hang out with Jensen Huang, CEO of @nvidia, and do a long-form podcast with him. Really fun & fascinating technical deep-dive conversation on & off the mic. One of the most brilliant & thoughtful human beings I've ever met. NVIDIA is the most valuable company in the world by market cap and is the engine powering the AI revolution. Podcast probably out tomorrow (Monday) unless I get stuck in too many interesting conversations while running around in SF ;-) PS: I haven't checked my messages in days. Sorry for slow replies 🙏 Trying to stay deeply focused at in overwhelmingly…

  Related

Pseudonymous writer; posts careful, funny investigations of everyday questions — air quality, seed oils, arguing, chess — usually by reading the studies.

• The modern formatting addiction in writing

  Editor's note: At this point, the author became agitated and wrote and then deleted a bunch of bullet points. In the interest of transparency, these are collected here. (Click to expand.)

  AddictionRelated

British software developer at Laravel and creator of the Cachet open-source status page system. Podcaster; writes at james.brooks.page.

• tweet.new — Say hello to tweet.new, your shortcut to posting faster on X!

  Related

CEO of Vercel; creator of Next.js and Socket.IO. Writes at rauchg.com.

• ray-city — A city animation that fits in a tweet

  Related

Writer and engineer behind the blog Melting Asphalt, and co-author with Robin Hanson of The Elephant in the Brain.

• I'm excited about the LessOnline festival! May 31 – June 2 (Link in next tweet, smh Elon.....) Hit me up if you're planning to attend — it'll be great to see some internet friends IRL

  Related

Writer and engineer behind the blog Melting Asphalt, and co-author with Robin Hanson of The Elephant in the Brain.

• mind-expanding short story by @etirabys about a "sapient mitotic species" this is a rare piece of fiction that's well grounded in evolutionary biology and psychology i read it about 6 months ago and have re-read it twice since 👇 link in next tweet

  Related

Research scientist at Anthropic; co-creator of the Variational Autoencoder and, with Jimmy Ba, of the Adam optimizer, two of the most widely used tools in deep learning.

• Funny how the replies reveal that many people can't actually shape rotate in their mind
  Quoting@ylecunTo people who claim that "thinking and reasoning require language", here is a problem: Imagine standing at the North Pole of the Earth. Walk in any direction, in a straight line, for 1 km. Now turn 90 degrees to the left. Walk for as long as it takes to pass your starting point. Have you walked: 1.…

  Related

Brazilian-American staff software engineer working on platforms, AI and developer productivity. Writes about engineering and his setup at ivansantos.me.

• That's my first tweet with @typefully. How come I didn't know about this tool before? 🔥

  Related

Co-founder and partner at Union Square Ventures in New York. Has written the AVC blog almost every day since 2003, on venture capital, startups, music and crypto.

• This tweet stream showcases some of the writing by USV analysts over the years. Though they have moved on to bigger and better things, their writing and their ideas live on at USV and https://t.co/btPOKgjAa4. It is a gift that keeps on giving.
  Quoting@nickgrossmanReminder that applications are now open for the next class of @USV analysts! One of the ways that USV analysts work is through writing, both about thesis areas and new investments. To give a flavor, here are some highlights of writings by USV analysts over the years:

  Related

Co-founded Neuralink and founded Science Corp, which is building a retinal implant; writes about neural interfaces and biology.

• What does Twitter asymptote to?

  Consider the following interpretation of Twitter: Twitter is a process that is applied to a human thought vector $x$, a time $t$, and a state vector $c_{t-1}$ which emits a value $tweet$ and a new state vector $c_{t}$.…

  Related

Engineering leader and writer; wrote The Manager's Path and Platform Engineering, and blogs at Elided Branches.

• Driving Cultural Change Through Software Choices

    This tweet got me thinking about change, and how software engineers (and especially, Platform teams) can drive cultural change throughout companies. First, let’s take the question. You want to change the engineer…

  Related

Founding member of OpenAI and former director of AI at Tesla; creator of nanoGPT and the term "vibe coding".

• A Recipe for Training Neural Networks

  Some few weeks ago I posted a tweet on “the most common neural net mistakes”, listing a few common gotchas related to training neural nets. The tweet got quite a bit more engagement than I anticipated (including a webin…

  Related

Financial journalist; chief financial correspondent at Axios and host of the Slate Money podcast, previously a blogger at Reuters and author of The Phoenix Economy.

• Nota bene: Gallows humor

  Did you laugh when you saw this tweet?

  Related

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 Ezra Klein Show: VC Bill Gurley on Transforming Health Care

  In November of 2015, I posted a tweet that declared Benchmark was interested in discovering Internet healthcare investments. Our firm has had the good fortune to invest in many two-sided networks that used information a…

  benchmarksRelated