AI slides but they actually don’t suck 🙏🏽
Replit ⠕: You're going to be embarrassed by the slides you made before AI
Meet Replit Slides
The first AI slides with stunning design

Peter has had some awesome podcast episodes lately. A lot of insight into creative and progressive ways people are building with AI.
Peter Yang: "I shipped 9 failed products before one took off...now I'm doing $1M+/month."
Here's my new episode with @tibo_maker, a solo founder who bootstrapped 5 AI products to $1M+ / month.
Tibo walked me through his exact playbook:
✅ How to validate ideas and fail fast
✅ Why his top

MIT student asked a question earlier today that a lot of young founders are quietly wondering about:
"Won’t the frontier labs just do everything?"
Yes it's true that OAI/Ant are shipping at incredible pace, but it's quite easy to avoid their blast radius and build amazing startups:
OpenAI is not going to build a cattle-herding drone, buy an old F-150 and drive from ranch to ranch like the founder of one of the fastest-growing YC W26 startups, Graze Mate.
Anthropic is not going to integrate with dental insurance verification systems (Lance).
Google is not going to navigate NATO procurement (Milliray).
The value is in the last mile, not the model. Sales cycles require humans who understand the customer. And most importantly, the market is expanding, not shrinking: AI isn't cannibalizing the existing 1% software spend — it's unlocking the other 5-6% that was going to humans. That's a much bigger market for startups yet-to-be-founded than the one the labs are playing in.
Now, what DOES seem risky?
A thin UI layer on top of ChatGPT with no domain expertise; a general-purpose chatbot or assistant; or a product that gets obsolete when model capabilities improve.
But — tools for specific industries; "full-stack" AI companies that actually are the service (AI law firm, AI accounting firm, AI uranium exploration company); or generally products where the customer doesn't want a tool but an outcome — are defensible ideas for startups.

This is actually brilliant for personal agents @Shpigford - stealing this for my OpenClaw now.

Can AI Dance? 🕺
Testing @Replit for slide creation
Replit ⠕: You're going to be embarrassed by the slides you made before AI
Meet Replit Slides
The first AI slides with stunning design

found a skill combo that works in both codex and claude code: /ce-ideate (from Every's compound-engineering) + gstack-design-consultation (from Garry Tan's gstack), running together. great way to build a new feature.
example: building a sticker chart app for my kid. sweep her room, earn a star. ten stars trade for a real reward.
/ce-ideate lists 30 directions, kills 25, keeps 5 worth shipping with reasons why each survived (does the parent phone show progress? do we add a leaderboard?). gstack-design-consultation runs in parallel, brings a designer's eye into the loop.
after both finish, i'd chain /ce-plan or gstack's /plan-eng-review to turn it into an execution plan with a task list. drop the execute command into codex's queue, queue a /review behind it. want to audit the design too? /design-review runs alongside /review in parallel.
then go grab food or take a walk. come back, PR is sitting there ready to merge.
ideate + design → plan → execute → review. and honestly you can queue the whole chain if you're being lazy lol.

"A decade ago, AI was supposed to replace radiologists. Today, radiologists make more than $500,000 per year, and their employment continues to grow, see chart below. Reading scans is a task, not a job, and when the task gets cheaper, demand for the job grows."

Just went to visit Legora. Most impressive startup I've been to visit in years. They're going to surpass Harvey in 2027. After that their only potential rivals will be the model companies. And if ever there was a territory you could defend against the model companies, law is it.

terminal has been my primary interface to my computer for almost two decades. now it’s the Codex app.
Yam Peleg: I was not expecting the Codex App to be even better than using the terminal.
Highly recommend everyone to try.
If you are on Linux just tell GPT-5.5-xhigh to “find a way to get it, it’s known to be easy”

Demis Hassabis (@demishassabis) has had one of the most extraordinary careers in tech.
He started as a chess prodigy and video game designer at 17 before getting a PhD in neuroscience and going on to found DeepMind. His lab cracked Go, solved protein structure prediction with AlphaFold, and then gave it away free to every scientist on earth. That work won him the 2024 Nobel Prize in Chemistry. Today he leads @GoogleDeepMind, pushing toward the same goal he set as a teenager: AGI.
On this special live episode of How to Build the Future, he sat down with YC's @garrytan to talk about what still needs to happen to get us to AGI, his advice for founders on how to stay ahead of the curve, and what the next big scientific breakthroughs might be.
01:48 — What’s Missing Before We Get To AGI?
03:36 — Why Memory Is Still Unsolved
06:14 — How AlphaGo Shaped Gemini
08:06 — Why Smaller Models Are Getting So Powerful
10:46 — The 1000x Engineer
12:40 — Continual Learning and the Future of Agents
13:32 — Why AI Still Fails at Basic Reasoning
15:33 — Are Agents Overhyped or Just Getting Started?
18:31 — Can AI Become Truly Creative?
20:26 — Open Models, Gemma, and Local AI
22:26 — Why Gemini Was Built Multimodal
24:08 — What Happens When Inference Gets Cheap?
25:24 — From AlphaFold to the Virtual Cells
28:24 — AI as the Ultimate Tool for Science
30:43 — Advice for Founders
33:30 — The AlphaFold Breakthrough Pattern
35:20 — Can AI Make Real Scientific Discoveries?
37:59 — What to Build Before AGI Arrives

OpenAI vs. Anthropic gets all the attention, but @GoogleLabs is quietly shipping some of the most interesting AI products right now.
I made a 16-min video showing you what 5 Labs products are capable of:
1. Pomelli for marketing
2. Stitch for design
3. Genie for 3D worlds
4. Flow for videos
5. NotebookLM for synthesis
📌 Watch now: https://youtu.be/BQ_mgojj0Mo

Noetik’s AI value proposition: “cohort selection”
“When I do my drug’s clinical trials, who should participate?”
Obviously, if you get that wrong, then your drug won’t “work,” even if it actually works!
Ron Alfa explained to us how they do this using cheap imaging
1/

Full Workshop: @OpenAI Codex masterclass
The agent is no longer just one chat window. In this workshop, @reach_vb and @kagigz get into how coding systems start to change when you can delegate work across subagents, split tasks up, and manage more context than a single thread can comfortably hold.
If you're trying to understand where the next generation of coding agents is going, this is a useful look at the shape of that future from inside @OpenAI!
Full Workshop: https://youtube.com/watch?v=MhHEGMFCEB0
Vaibhav (VB) Srivastav: http://x.com/i/article/2033608358494617600

i miss @AIEMiami already

Expand is now available to everyone!
Extend your world in any direction you choose: around corners, into rooms, and beyond what you can see 👀

Building apps is easy- keeping them running isn’t
Introducing Replit Application Monitoring
Replit Agent now watches your app in production, investigates issues, and helps fix them- so you don’t have to

Proactive agent that thinks and acts like you.
Multiplayer AI Brain for teams.
Proper GUI for commanding 50 agents.
http://Sauna.ai is all three. Sauna goes live today. First 2000 people, use access code LAUNCH for $80 of weekly(!) credits.
Let’s explain. Multiplayer only works once the personal brain is powerful. So let's start here.
Personal AI Brain
3,800+ tools connected. State of the Art memory. Skills and schedules you teach once that get repeated forever on cheaper models. An AI first CRM.
Lives on the cloud so you can initiate tasks from anywhere: iMessage, Slack, Email. Also no need for a Mac mini 😉
GUI
AI agents have been stuck in their MS-DOS era. A chat box, a scroll buffer, no way to command 50 of them. We built the first GUI: Live sessions on one side, work waiting for your sign-off on the other, plus the things Sauna kicked off while you were asleep waiting for review. Game mode helps clear the queue with actual joy.
Okay so far so good, but how to give benefit of what you built to more people or whole team?
Multiplayer
Once your Sauna actually knows you and you gave her access to your tools, you can use multiplayer.
Two modes:
- Brain access. My co-founder Robert plugged my brain as a tool into his Sauna last month. He can ask it about pricing while I'm in other meetings, gets a sourced answer back, never has to interrupt me. That’s read only. Yolo mode gives him access to all my tools too :O
- Communal Saunas extend that to whole companies with proper permissioning. Folder owners decide what's true for the whole company and build skills. Most get their personal brain + read access to communal files and memories. Works also for group planning my best friend's bachelor party.
The Way
I’ve been obsessed about AI Brain since 2016. Our human brains suck at some things like memory and are brilliant at others like creativity. We are also particularly bad at thinking we are all on the same page and then realising weeks later that we weren’t. Most leaders spend their days being the human diff tool, catching contradictions in hallway conversations and Slack threads. Repeating themselves 50 times. Now every company is spinning up hundreds or thousands of agents that drift faster than humans do, feeding each other their drift as context. Compounding rot.
After 10 years and one failed company in this space we are launching the solution.
The Launch
We thought about a celebrity launch. Margot Robbie in a bathtub explaining agents. But then we realised the same money gives the first 2,000 people free daily credits, every day, until we burn through that $1,000,000.
Sauna runs Claude, GPT-5, Gemini, GLM-5.1, DeepSeek, Kimi so you can budget yourself. The labs are racing to lock you so they can milk you in a year. We picked your side.
Onboarding
It’s live at http://app.sauna.ai We’ve preheated saunas based on a niche. Use the access codes in the comments to get a better experience.

Pitching deck for investors under $1.3
Magomed Kurbaitaev: I swear I spent over $10,000 on pitch decks when I started Gameplan.
Agencies, designers, revisions, back-and-forth, more revisions and “final_final_v7.pdf”.
Replit just built an investor deck in 4 minutes for $1.23
I am still processing this. Insane.

Did a very different format with @reinerpope – a blackboard lecture where he walks through how frontier LLMs are trained and served.
It's shocking how much you can deduce about what the labs are doing from a handful of equations, public API prices, and some chalk.
It’s a bit technical, but I encourage you to hang in there - it’s really worth it.
There are less than a handful of people who understand the full stack of AI, from chip design to model architecture, as well as Reiner. It was a real delight to learn from him.
Recommend watching this one on YouTube so you can see the chalkboard.
0:00:00 – How batch size affects token cost and speed
0:31:59 – How MoE models are laid out across GPU racks
0:47:02 – How pipeline parallelism spreads model layers across racks
1:03:27 – Why Ilya said, “As we now know, pipelining is not wise.”
1:18:49 – Because of RL, models may be 100x over-trained beyond Chinchilla-optimal
1:32:52 – Deducing long context memory costs from API pricing
2:03:52 – Convergent evolution between neural nets and cryptography

Democrats could try picking a young candidate who has a solid record of running a big city https://www.economist.com/united-states/2026/04/29/a-radical-idea-for-governing-california?taid=69f249c405fd1e0001605013&utm_campaign=trueanthem&utm_medium=social&utm_source=twitter

Re There are so many shady, suspicious things about the ethnic studies program:
1) SFUSD mandated it to be 2 semesters, but CA law only requires 1 semester. That other semester takes away additional instruction in math, english, science, etc..
2) Parents can't preview the content in "Voices An Ethnic Studies Survey". Why keep it secret? They say it's for copyright reasons, but even Amazon often allows users to preview many pages of a book online before buying. That especially should be important for parents concerned for their children.
3) the company that publishes the book also hides who is behind it. Their website & "about" section isn't even transparent enough, to show a single person's name. https://gibbssmitheducation.com/about That's unusual for a company website, and it should be doubly so for a company that serves children.
4) People commonly know what to expect with other classes: Geometry, World History, Social Studies -- what's inside is generally standard. But "ethnic studies?" It's more of a mystery. It's a recently made up subject for high school. And it is highly prone to politically biased & manipulative content.
Too much secrecy & hiding. Too much rushing through.
Overall, way too SHADY.

This is what an impending midterm disaster looks like for Republicans. https://trib.al/eLV1GmR

I mentioned Liquid AI to @swyx - and he immediately got Maxime to discuss their approach to training. Worth a listen.
swyx 🇸🇬: @MParakhin @ilyasut and first AIE x @liquidai talk is live!
https://x.com/aiDotEngineer/status/2049458425059537025

Seems illegal
Also this is why people hate progressivism
Aaron Sibarium: NEW: Blue jurisdictions are rationing homeless services based on race.
In Portland, a non-white, non-native English speaker who is LGBT would get priority over a domestic violence survivor with a 6 yr old child who's been homeless for 12+ months.
The policies are shocking.🧵

Re 6/ Most "AI for business" companies right now are still operating at Layer 1 with prettier UIs.
The interesting work this year is Layer 3.
@garrytan keeps pointing at the right shape with GBrain and GStack. The companies that figure out how to compile a brain — not just index one — own the next category.

Science is how we’ve cured diseases and how we just sent four humans around the Moon. It’s a big reason why we’re the greatest country on Earth, but firing the National Science Board shows just how much Trump doesn’t understand that.

Re @vercel Ship26 is live!
This year, we shipped a 3D device that can create WASM apps using wterm, just-bash, and @workflowsdk
...and a landing full of agents that walk around the site using a dynamic navigation mesh.

Thanks @mattshumer_ for finding one more gem use-case for the incredible GPT-Image-2. You can generate stop motion videos just by asking !
https://x.com/mattshumer_/status/2049583863576617223?s=20
Matt Shumer: GPT-Image-2 can be a video-gen model if you want it to be!
Just ask ChatGPT to generate stop motion video, frame-by-frame.
It works shockingly well and is super controllable!

Un truc qui m'a marqué en arrivant à YC.
Premier meeting, @garrytan nous dit : "dans cette pièce, 10% d'entre vous vont potentiellement devenir milliardaires. Faites-vous des potes avec tout le monde. Parce que même si votre boîte ne marche pas, vous allez pouvoir vous entraider toute votre vie."
Et c'est exactement ce qui se passe. Il y a une bienveillance vraiment rafraîchissante dans cet écosystème. Personne ne te regarde de travers parce que tu vises trop haut. Au contraire, plus tu vises haut, plus les gens veulent t'aider.
C'est ça qui manque cruellement à la France.
Le soutien à l'extrême ambition.
Tant que dire "je veux devenir milliardaire" sera un problème social en France, on n'avancera pas. Tant qu'on n'aura pas de bienveillance écosystémique sur ces sujets là, on stagnera.
Le blocage français n'est plus technique, ni même réglementaire au premier ordre. Il est culturel. C'est un blocage sur la création de valeur elle-même. On a intériorisé l'idée que la richesse est suspecte, que l'ambition est vulgaire, que viser grand c'est trahir une forme d'humilité républicaine.
Pendant ce temps, on rentre dans une décennie d'abondance extrême. L'IA, l'énergie, la robotique, la biotech. Les ordres de grandeur de création de valeur vont être historiques. Et la question n'est pas de savoir si ça va arriver, c'est de savoir qui va capter cette valeur.
Si on continue à punir socialement ceux qui essayent, on regardera les autres le faire. Encore.
On a un vrai travail collectif à faire en tant que français. Réapprendre à célébrer ceux qui construisent. Arrêter de confondre ambition et arrogance. Comprendre qu'un écosystème où les gens s'entraident à viser haut produit plus de réussites pour tout le monde, pas moins.
La bienveillance envers l'ambition n'est pas naïve. C'est stratégique.

I asked Replit to turn my tweet into a deck
Replit ⠕: You're going to be embarrassed by the slides you made before AI
Meet Replit Slides
The first AI slides with stunning design

Trump’s FCC chair, Brendan Carr, says he plans to punish and potentially revoke broadcast licenses for TV networks critical of the Trump administration:
“I’d be surprised if we don’t go down the path of license revocation for some. It will be up to them and their conduct.”
This is fascism. This is what they do in authoritarian countries.
This is cancel culture on steroids.
Where’s the outrage?