Shawn Wang

RT AI.Engineer @ CODE 11,000 pull requests in one month. A revert rate around 0.5%. @poteto on the Grok Bot team's shipping pace at @SpaceXSI. More contributors, daily stable releases, and an app that got faster. From AI Engineer Code Summit.

also the day 1 livestream has crossed 100k in the fastest time ever. covering coding agent practices and landscape continues to be the strongest format we do and by far my strongest product-market-fit focus thing I have ever done.

RT Ahmad Attended an amazing broadway musical tonight, don’t sleep on @swyx outside the AI world dude is incredibly talented and full of surprises

holy shit u guys like a lot of poteto

holy shit u guys a lot of poteto

obviously correct, we are so suboptimal in AI UX rn it is crazy esp given existence of models too cheap to meter

RT Matt Stockton The more I learn about how the internals all work, the more I agree with this take. This week’s @latentspacepod / @swyx podcast with Handshake was super-interesting and eye-opening to me. Almost 3 hours long but worth the listen - and probably just scraping the surface of what all goes into this. Space still feels very early

if you're in SF, i have 3 spare tix for our matinee show, just show up to odc theater now and drop my name 2.30pm doors open 3pm starts

openai dot now has a setting to just solve all captchas for the rest of time

RT dex a few days ago we published the "No AI Slop Manifesto" at @aiDotEngineer Code Summit in SF, written in collaboration with many of the speakers and attendees It already hit the front page of hacker news, has 300+ signatures including folks from Baseten, SpaceX, OpenAI, PostHog, Shopify, Glean, ZoomInfo, and legendary builders including @swyx @poteto @kentcdodds @mihail_eric @audreyt @altryne @traviscline @philipkiely @TejasKumar_ read and sign it at slopmanifesto dot com some of my favorite points > AI cannot replace human judgement, it can only amplify our judgement (or lack thereof). > We own the intent, taste, and consequences of what we ship, however much AI helps us make it. > There will always be alpha in reviewing *something*. As models improve, we may stop reading most of the code. We may still review a plan, design, or result, because if we don't, we just get the same output as everyone else.

RT AI.Engineer @ CODE AI can do more of the work. So why does the job feel harder? @simonw on what happens when the easy tasks disappear and your ambitions catch up with the tools. From AI Engineer Code Summit.

RT Maz Pretty cool in San Francisco you can watch @swyx perform musical theater for a cool 34 bucks !!! Link to tickets below

RT Amit Vadi what a week at @aiDotEngineer. thanks to @swyx, @Beccalytics and team for putting so many world class builders in one room and to @melvinjohnsonp for closing things out on stage. shipping season is in full swing. see you at AIE Europe in Feb!

RT Henrique Bastos I had a great time at the after-party yesterday: good chats, new friends, big ideas, and many agents to kick off on the flight home. In the last conf photo: the amazing @miguelbranco80 and @dexhorthy. Thanks to everyone at @aiDotEngineer for the experience. See you soon.

RT Alex Volkov AIE code summit is done! @swyx said this was the 10th one! What a great event this is! Much higher quality folks, conversations are full of alpha, new ideas, zeitgeist updates is top notch! Also I heard X allows a photo dump now? Great for post event summary! This time besides doing the regular LIVE ThursdAI show, me and @RayFernando1337 ran around and did a hallway track stream, and talked to tons of speakers, attendees, and organizers. The event was impeccably organized yet again, this time with @Beccalytics and @g2i_ai great execution and taste! Food was dope, after parties and speakers dinners were superbly organized! Really my only complaint is WiFi but that's up to the hotel usually? But so many great folks are getting into the AIE circle, that it's more than compensates for the poor connectivity! Got to see old friends and make many many new ones, learn a ton about Factories from @dexhorthy @kentcdodds and @zachlloydtweets among others, formal verification from @poteto (also shared an actual poteto ha!) and type one models from @michellechen and @ivanleomk The three themes this time according to Swyx were: - Factories - Type One models - Dreaming bigger Saw an amazing crowd around @echelon - who represents the "dream bigger" at this event with ArtCraft 👏 And of course felt like a Billionaire hanging out with @DevinAI at the Token Billionaires lounge! (Huge shoutout to @mattbergland yes again!) If you feel FOMO - please do check out our coverage we had a lot of fun getting behind the scene and showing you the hallway track

and straight to broadway show rehearsal :) see you saturday

the production team got us a cake!! #aie1m

RT @Beccalytics: When @swyx asks you to set up a Smash Bros activation for @Steve_Yegge you make it happen. 45 mins later…thanks to @Judg…

RT Steve Yegge If anyone here at @aiDotEngineer thinks they can beat me, game on.

RT Hank Taylor Sometimes I miss being the man behind the curtain at dev conferences (Next.js Conf and Laracon), but attending events with no responsibility is a pure delight Wish I'd had more time to hang at @aiDotEngineer and catch up but had to cut it short (will explain in next tweet) @swyx thank you for creating these and getting cool people there, + @Beccalytics (and many many others) for impeccable execution! I've been hanging out with these folks for years now, fun talking to them about the radical shifts in our careers and industry and also just having friends in faraway places. And of course the people I spoke to the most are the ones I forgot to take pics with 🤦

RT Justin Poehnelt Re @swyx and team provide a great experience and many benefits to speaking at @aiDotEngineer. 🙏 The speaker swag/bling is nice too!

RT Jay Masilamani I always knew @jxnlco is an applied philosopher since I first took his class. Iconic talk today! In a doom and gloom climate, many are searching for answers on how their careers will unfold. If that is you, go watch his talk - the first one on the stream. https://www.youtube.com/watch?v=y9JtCCZP2MQ My takeaway is this: make space for exploration and learning. Great pick @swyx, to make this the first talk of the day to set the tone.

RT Nader Khalil🍊 This is a difficult moment. It’s hard not to look at the journey this company has taken and not be filled with respect for what Jensen built. The spirit of the company remains strong: "mission is the boss”. That said, as NVIDIA finds its way forward, I want one thing to be clear: NVIDIA's contributions to open source are unwavering. Jensen has been vocal about this, and that commitment runs deep in NVIDIA's people and culture. We all understand the importance of open source. NVIDIA is a huge supporter of open source because we understand something. If you want to diffuse technology, you do so through an ecosystem, not a single company or entity. No one owns the internet, no one owns servers, and no one owns operating systems… but through open source, an ecosystem of closed and open innovation flourishes. When we talk about delivering agents safely and securely, that too will take place because of open source. For example, ensuring people have the best models to secure themselves at machine speed unburdened by arbitrary guardrails decided behind closed doors. Open source & AI are often discussed together, but it’s worth reminding, open source predates AI. It is a mechanism for sharing ideas inspired by the hippies at computer labs of California in the 1960s. Before AI, the largest shift I lived through was the internet. It was remarkable seeing the fight for open source take place there. Between browsers, runtimes, net neutrality, etc. What came as a result was HTML5 and an extremely vibrant ecosystem that built the internet as we know it today. NVIDIA and its partners hold an incredibly important role in delivering this technology. NVIDIA contributes to every layer of the stack. Runtimes, models, harnesses, programming languages, frameworks, etc. None of that changes. Long live open source.

“if you want good taste, you have to eat” - @jxnlco

Apple: “Introducing the new Siri” Me: “oh sick, can you turn off all my alarms” new Siri:

Kent helped me get started in my second career and it is such an honor to host him as ai engineer evolves into product engineer! def keep an eye out for https://kody.codes, this category has not yet been created but it is about to get some serious firepower behind it

RT Latent.Space The System One revolution is upon us! At @aiDotEngineer Code Summit today, lead Jev creator Diogo Almeida (@CompleteSkeptic) called on developers to automate everything. Also, Glean and others showed how they’re using Jev. https://latent.space/p/jev-revolution/

RT Josh Birdwell What an incredible first day at AI Engineer Code Summit. Heard some amazing talks! I want to shout out a few of my favorites: @poteto @CompleteSkeptic @Steve_Yegge @simonw Some of the themes: let's be more ambitious. Be intentional with building trust in your agents. Finally got to meet my friend @thecsguy IRL today too. @swyx and team, thank you again for putting on the conference. Can't wait for tomorrow.

RT Turing Post (Ksenia Se) “Software is gonna be so fucking back.” .@CompleteSkeptic is full of bangers and you will have to listen to his full talk at @aiDotEngineer, but for now, here is our top-10 of his pearls: 1“We’ve been announcing AGI three months away for two years now.” Meanwhile, go outside San Francisco and a lot of the world looks unchanged. Where the fuck is all the automation? 2“You get Einstein 95% of the time, but Mr. Bean 5% of the time.” And if you can’t tell which is which, how are you supposed to automate anything? 3“I do believe RLHF is part of the problem.” His complaint: we optimized models for assistance, with a human checking the result, and then expected them to reliably do work on their own. 4“The bitterest lesson is doing the right task.” Diogo thinks that compute matters more than algorithms, data matters more than compute, and choosing the right task matters most. You get what you optimize for. 5“I think the world still has LLM Stockholm Syndrome.” We keep designing around the limitations of today’s models. Why is the best we can come up with still the same software, cheaper to write, with a chatbot on the side? 6“What if we can program with common sense, with intention, with human understanding?” Very passionate talk! He suggests: put intelligence into the software’s inner loop. What could we build if understanding intent became a building block we could actually rely on? 7“Real fucking engineering by people who care a lot about the end result.” Break tasks down, build stages you can verify, and actually manage state. Why should every agent rediscover the same codebase patterns over and over? 8“There’s an asymmetric advantage for people who are, like, bad actors and don’t give a shit.” That’s his explanation for slop: unreliable output is much easier to tolerate when everyone else pays the cost of reading, checking, and filtering it. 9“I call this software research. That’s not a joke.” Explore what software could become, including we...

this person is a real former employee that we parted ways with that email domain is not one we use

next on stream - @CompleteSkeptic and @typesafeai take over for Jev Day!

standing room only for @poteto’s keynote

RT Drew Bredvick Re @swyx amazing episode with @joinHandshake. Really practical behind the scenes look at the data industry and I think there’s a lot of gold in here. Go listen if you have a few mins today https://share.snipd.com/episode/9bc342b0-fe16-440a-ac60-35b86106db93

RT Alex Volkov These awesome ladies are killing it at @aiDotEngineer code! 👩‍💻 IIRC this is the first all female keynote start at AIE? @swyx ? Back to back: @nserrino - 10X inference @lydiahallie - @claudeai best practices @michellechen - Cloudflare being a neolab @poteto - poteto poteto

livestream starting now!!

the bigger you are, the bigger the target on your back. cyber security is now job 0 of every ai engineer

i had a preview of his keynote and you guys don’t know what youre in for in thursday. literally had to agree to embargo to get this talk on board (it is not the full story) tune in to the aie live stream 9am-6pm tmr and thursday!!

RT Alex Volkov Early vibes at AIE are dope! Excited for tomorrow! As always an incredible group of folks, and knuckleheads @aiDotEngineer

RT Turing Post (Ksenia Se) Re Super cool stickers @swyx !!!

RT Turing Post (Ksenia Se) Hot AGI pilled at @aiDotEngineer ! Say hi if you are here! And if you are not watch all the talks online - content is always incredible

RT Ahmad Thanks for everyone that showed up to my talk and sorry for the 170 slides 😂

I finally asked @zenorocha, the CEO of AEO, why tf my agents keep choosing Resend even when I don't want them to

prediction: tech jesus in 10,000 years

RT Latent.Space 🆕 Designing for Agents Agent Experience is one of the most hotly contested topics of 2026, and increasingly critical for devtools/infra companies to get right. @ZenoRocha has been writing and thinking about this since 2024 and came on the pod to spill @Resend secrets! https://youtu.be/bdIoODXsu8k Key themes include: - Answer Engine Optimization (AEO): how AI models like Claude Code recommend email services and how product teams can design for these agents - Deliverability and Trust: The discussion delves into the complexities of getting mail into the inbox, including handling bursts, managing Gmail throttling, and the importance of domain reputation and anti-abuse measures - Developer Experience & MCP: A major focus is on building for both humans and machines. Resend utilizes the Model Context Protocol (MCP) to provide agents with over 80 named tools, allowing them to autonomously set up accounts, check pricing in Markdown, and send emails - Technical Infrastructure: React Email as a wedge for adoption and but Resend's commitment to owning the mail stack, including managing complex DNS hurdles and providing enterprise-level visibility

"If you're coding locally, stop. You really need to be coding with a cloud agent." notes of @adamwathan and @ScottWu46 in this one

not kidding. scenes from the speaker rehearsal session last night. @lucasmeijer , @FabianHedin, @BuiltByJustin, and @kentcdodds made memorable splashes!! follow @aiDotEngineer for live tweeting

RT dex if you are at @aiDotEngineer today DO NOT miss @lucasmeijer ‘s workshop today it is an full hour of straight gold - pragmatic, tasteful tricks, prompts, workflows, and general advice

RT Kent C. Dodds 🐨 This is where I'm talking about turning your codebase on autopilot. See you at @aiDotEngineer at 11:40!

bro just proved fermat's last theorem AS A DEMO for the fable 5.2 launch and casually told me he's got something bigger planned for his aie code talk lol anthropic u guys are wildin

btw by complete coincidence the $ANT IPO roadshow across wall street started today. they had this in the chamber and its also nice that it came after Astra, Argon and Carbon…

ok i was signed in to wrong account to tweet this but yes - if you're looking to get a start with technical writing/speaking/tweeting/blogging, the new Creator associate program is made for you... ... just like I got started in my own technical writing journey 10 years ago.

RT Mihail Eric What do you teach computer science students when an agent can one-shot all of their homework? I sat down with @swyx on @latentspacepod to walk through how I'm answering that in my Stanford course 𝗧𝗵𝗲 𝗠𝗼𝗱𝗲𝗿𝗻 𝗦𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿, and how the same ideas show up day to day at @MonacoGTM. Last year all students were just using tab complete. This year when I surveyed them, every one was regularly using autonomous agents. Exercises that once taught the workflow were starting to feel one-shotable. So the curriculum shifted toward judgment: 𝗕𝘂𝗶𝗹𝗱 𝘁𝗵𝗲 𝗵𝗮𝗿𝗻𝗲𝘀𝘀: students write a small coding agent from scratch (about 200 lines), then intercept the harness of a state of the art coding agent and read the system, plan, and compaction prompts. To effectively use and build agents you need to understand the loop they run in. 𝗟𝗲𝗮𝗿𝗻 𝘀𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝘁𝗮𝘀𝘁𝗲 𝗯𝘆 𝘀𝗵𝗶𝗽𝗽𝗶𝗻𝗴 𝘁𝗼 𝗿𝗲𝗮𝗹 𝘄𝗼𝗿𝗹𝗱 𝗰𝗼𝗱𝗲𝗯𝗮𝘀𝗲𝘀: part of the homework is contributing to open-source projects. Maintainers review the patch and help students build mental models of good software. An agent can draft the change but getting it accepted still requires appropriate context, architecture, and intent within a complex codebase 𝗨𝗻𝗱𝗲𝗿𝘀𝘁𝗮𝗻𝗱 𝘁𝗵𝗲 𝗰𝗵𝗮𝗻𝗴𝗲: before implementation, research an unfamiliar repo, propose a few approaches, and choose one. After the agent writes code, quiz yourself on the diff until you can defend it without reopening the file. At Monaco, that judgment is most of the job. When agents can build almost anything, the hard call is what's worth building. We buy tools to learn a space, then build only the pieces that have to fit our own infrastructure. Code review has stayed a bought product for a while (though that's starting to change!). Agent workspaces, where background agents run the app and test patches, we built ourselves. In a rapidly evolving space, staying nimble and adaptable is one of the few de...

starting a thread of all the ways in which gmail sucks for the eventual day i vibecode a total replacement

good read on why humans, like llms, get a lot more powerful when given a scratchpad.

are you listening anon?

RT Ryo Lu when we can do anything, do we still know what we want? the full talk from @aiDotEngineer shanghai. every slide handmade in keynote, subs in 中文 and english

RT @vtahowe: @swyx Let's gooo!!! Big congrats on 1 million! 🎉 My favorite is any debate with @ianlivingstone and @dexhorthy https://t.co…

RT dex working on something fun w/ @swyx and @aiDotEngineer team - i like to think at least some of y'all have very good taste. Please share your thoughts: (obvs i bias toward code but this is important broadly!) What does slop mean to you? Either in terms of code slop or more generally Why does slop happen? Why is slop worth fighting? What happens if we don't prevent slop? What happens if we do? How can we fight and prevent slop?