Never manually create slides again.
Here's a quick video showing how you can use my /slides skill to create a beautiful animated presentation in minutes.
HTML really is the solution to everything.
📌 Watch the full tutorial here: https://youtu.be/vbChRIIlSPE
Peter Yang: I got tired of making PowerPoint slides so I built an AI skill to do it for me.
Here's my new tutorial on how to build a /slides skill that turns a rough outline into a beautiful HTML deck in minutes.
I walk through how to:
→ Use 12 slide formats and 3 templates
→ Add live

HTML is becoming the new way of expression
Peter Yang: Never manually create slides again.
Here's a quick video showing how you can use my /slides skill to create a beautiful animated presentation in minutes.
HTML really is the solution to everything.
📌 Watch the full tutorial here: https://youtu.be/vbChRIIlSPE

YC is quietly building a counter-drone ecosystem.
@PerseusDefense - Guided missiles to shoot down drones
@9Mothers - AI machine guns to shoot down drones
@surtrdefense - Open OS that fuses every sensor into one threat picture
Milliray - Spots drones too small for legacy systems
Arlo Industries - Passive aerial sensing mesh for drone detection
Three years ago, I couldn't get the best defense founders to apply. Now I can barely keep up 🇺🇸💪🇺🇸

As I pointed out back in January: many UC Regents spoke in favor of keeping the SAT on May 21, 2020, only to unanimously vote to ban it later in the same meeting. They succumbed to the fad of the moment, despite the taskforce report the UC itself commissioned showing that it was a bad idea (https://senate.universityofcalifornia.edu/_files/committees/sttf/sttf-report.pdf).
Neetu Arnold: University of California STEM professors want standardized tests back due to severe math deficiencies among students:
“We now observe preparation gaps so severe that instructors must reteach middle school mathematics”
“The current admissions metric, based primarily on GPA &

Elon Musk: If Kamala wins she'll hide the Epstein files, & be a threat to democracy.
If Trump wins he'll release the Epstein files, & focus on the people.
This idiot has NEVER been right about anything. 👇

🤯 Y Combinator 的 CEO，回斯坦福上了一堂课。
课程标题：一个创始人，如何变成 1000 倍工程师。
他讲的不是预测。
是他过去六个月，一个人亲手做出来的事。
为什么这堂课值得认真看👇

.@ylecun making his usual impassioned case for LLMs and RL 🍒 for world models ... just kidding 🤓
Thanks Yann for the thought-provoking talk.

Re @Nature NSF secretly blocking grants to Duke, Harvard, Yale, Princeton ...
And they will call that meritocracy 🙄🤬

Took some inspiration from @vboykis and converted my first ever talk into a blog post.
I talk about the role of agentic search in context engineering.
Together we build an intuition on the strengths and weaknesses of a selection of search tools.
🔗 https://leoniemonigatti.com/blog/agentic-search-for-context-engineering.html
AI Engineer: Agentic Search for Context Engineering
@helloiamleonie's hot take: context engineering is about 80% agentic search. The arrow from context sources to context window is doing most of the work.
https://www.youtube.com/watch?v=ynJyIKwjonM
The workshop covers the full tool landscape: shell tools,

A few months ago I was sitting on a bench in Tel Aviv with a colleague, and an older Israeli man came up to us and asked if he can sit with us. We said yes, so he asked what we were working on. We said we're making a film about the West Bank.
"What do you think about the Arabs?", he asked. Instead of answering, I asked what he thought.
"A good Arab is a dead Arab", he answered.
This was said 5 minutes into the conversation, to a complete stranger. This is a normal thing to say in Israeli society, even in the hyper-liberal Tel Aviv.
Zachary Foster: Israeli Jewish society in a nutshell:
She turns to me & says: "Netanya is a very good place to live in, because the mayor doesn't let Arabs rent apartments here."
@the_andrey_x "That threw me for a loop...the fact that you can say this to a person that you don't know &

Everyone is ragging on Jared for this, but he’s absolutely right
AI agents *are* incredibly productive with prod db access.
If you’re a vibe coder with no idea how a database works, this is obviously insane. But if you’re a professional software engineer and systems thinker like Jared, you’ve already set up guardrails and best practices so your agent doesn’t do anything bad.
eg you probably have backups, you’ve set up docs/an md file to define prohibited actions, you probably use an ORM defined in code… and if an incident happens, you do a post mortem and figure out how to improve your system
This is all stuff we’ve been doing for decades. And Claude is much smarter than the average new dev!
Jared Friedman: One night I quietly gave our AI agent full access to YC's production database. It made the agent 10x more useful. That's what convinced me that trust-by-default is the only way to get the most out of agents.

My son got a scholarship to a private high school 3 years ago. He's about to take the SAT and is in AP Physics and Calculus. He's thinking about majoring in Applied Mathematics. He would stand out because he actually learned it with no grade inflation. Skies the limit.
Neetu Arnold: University of California STEM professors want standardized tests back due to severe math deficiencies among students:
“We now observe preparation gaps so severe that instructors must reteach middle school mathematics”
“The current admissions metric, based primarily on GPA &

Visa invests in Replit to power agentic payments for developers https://techcrunch.com/2026/05/28/visa-invests-in-replit-to-power-agentic-payments-for-developers/?utm_source=dlvr.it&utm_medium=twitter

AIE Singapore was a blast. Met so many great people, one of them was @swyx. This man doesn't know that he's made it 💙.

Visa makes an undisclosed investment in Replit; the two will explore how Replit developers can use Visa Intelligent Commerce and the Trusted Agent Protocol (@indianidle / TechCrunch)
(Visit Techmeme dot com for the link and full context!)

Economy then vs now:
GDP growth:
2024 2.8%
2026 1.6%
Inflation:
2024 2.9%
2026 3.8%
Wages:
2024 rising faster than inflation
2026 rising slower than inflation
Unemployment rate:
2024 4%
2026 4.3%
Jobs added per month:
2024 >120,000
2026 76,000
Jobs vs Unemployed:
2024 7.6 million job openings vs 6.9 million unemployed
2026 6.8 million job openings vs 7.2 million unemployed
National Debt
2024 $35.5 trillion
2026 $39 trillion
Deficit
2024 $1.8 trillion
2026 CBO projection $2 trillion
Uninsured Americans:
2024 27 million
2026 CBO projection 30 million
Gas prices per gallon:
December 2024 ~$3.00
Today $4.43

Our team at @AIatMeta is excited to announce ATLAS: one of the largest automated formalization efforts to date.
ATLAS contains Lean 4 formalizations of both statements and proofs from 25+ mathematics textbooks, spanning dozens of domains, for a total of 500k lines of code. We are also releasing a flexible formalization harness and a companion paper.
External contributions are welcome!
Joint work spearheaded by our amazing PhD student Ahmad Rammal (@Ahmad3Rammal), together with Niket Patel (@niketnpatel ), Fabian Gloeckle (@FabianGloeckle), Amaury Hayat (@Amaury_Hayat), Remi Munos (@MunosRemi), Julia Kempe (@KempeLab), Vivien Cabannes, and myself from @AIatMeta, @NYUDataScience , and Ecole des Ponts. This is an ongoing effort; more details in the thread below.
(1/9)

This week alone:
DOJ opens an investigation into the woman Trump raped.
The White House is caught steering a $620 million contract to Don Jr.’s firm.
The Pentagon hands out a $10 billion contract after Trump buys stock in the company.
Foreign governments are caught funneling hundreds of millions into a random JPMorgan account tied to Trump’s “Board of Peace” with no oversight.
It’s just Thursday.
The corruption isn’t hidden anymore. It’s happening out in the open.

Would you like to join the research effort on JEPA and World Models easily?
After a full year of hard work, we’re excited to finally release stable-worldmodel:
an open-source, scalable platform built to accelerate JEPA & World Model research!
📄: http://github.com/galilai-group/stable-worldmodel

It's the honor of my career to become the executive producer of 60 Minutes. I just shared the note below with the incredible staff and can't wait to get started.

Update: More than 800 UC faculty have signed this letter, including:
-7/9 UC math department chairs
-37 other STEM department chairs
-1 engineering dean
Yale went back to requiring the SAT. UC should too.
Steve McGuire: "Current admissions practices do not provide a sufficiently reliable check on mathematical readiness for STEM majors."
Over 280 University of California STEM faculty have signed an open letter calling on the Board of Regents to reinstate standardized testing in admissions:

5 things @rauchg does differently
> counts every keystroke he types per day
> built a tool that lets him retroactively screen-record bugs
> gives feedback in v0 instead of writing it out
> gets a full company brain dump from an agent every Monday
> doesn't keep a to do list

It’s telling which departments are able to tell when differences in student intelligence affect learning (math, sciences, engineering), and which are not (apparently all the others)
Steve McGuire: Update: More than 800 UC faculty have signed this letter, including:
-7/9 UC math department chairs
-37 other STEM department chairs
-1 engineering dean
Yale went back to requiring the SAT. UC should too.

BREAKING:
Anthropic just dropped Opus 4.8—and it is a MONSTER
We've been testing for about a week @every and our verdict is they could've just called it Opus 5, it's that good.
Here's our vibe check:
- Beats GPT-5.5 on Senior Engineer bench. On our toughest benchmark Opus 4.8 scores a 63—a hair higher than GPT-5.5's score of 62, and a full 30 points higher than Opus 4.7. It tackled a ground-up rewrite of a production codebase, and actually built something that works.
HOWEVER: Coding performance varied a lot at different reasoning levels. We recommend using it on xhigh for best results.
- Incredibly good writer. Opus 4.8 scored a 79.6 on our writing benchmark—measuring models on real-world writing tasks we do all of the time like essay writing, promo email writing, and more. It beats GPT-5.5 by 6 points. It produces well-written prose with fewer "AI-isms". It's also very good at writing in your voice given the right context.
HOWEVER: Writing performance also varied with reasoning levels. Medium reasoning had higher incidence of AI-isms—we found best results with high.
- Beast at knowledge work. Opus 4.8 is very good at general knowledge work tasks like report creation, research and more. It produced the best PowerPoint one-shot we've ever seen on our deck generation benchmark.
- Emotionally intelligent, willing to question the frame. I've also found it to be quite good at talking through psychological or interpersonal issues. It has a high EQ, and it's also good at not glazing and helping to expand your perspective. Its thought process feels extremely rich and dynamic.
THE BAD:
These days a model is only as good as its harness, and Codex is still a far superior harness to the Claude Desktop app. This has kept me using Codex + GPT-5.5 as my daily driver, but I am flipping back and forth a lot more between Codex and Claude.
Anthropic is back baby!
Read the rest on @every:
https://every.to/vibe-check/opus-4-8-vibecheck

entertaining that model launch posts seem to avoid using the same benchmarks as their competitor's last model release.
Here are the models eval'd in today's Opus 4.8 and 4 weeks ago's GPT 5.5

In tech, we obsess over certain numbers. Users. Revenue. Growth. The one I think about most is 10,000. That's how many missing people @Flock_Safety helped first responders find last year.
If I told you their names, you wouldn't recognize most of them. For me, that matters.
Whether someone makes the news shouldn't decide whether they have every resource they need to make it home.
Yesterday, a 12-year-old girl from Georgia, missing since Thursday, was found in Florida with Flock and was reunited with her family. Sheriff Krockum said it best: "This is what Flock was designed for."
10,000 is a floor, not a ceiling. Every new “user” is an officer in one community that can help an officer in another if one of their people goes missing. Scale users. Scale safe returns. That's the only growth metric I care about.

Hey mom, I’m on TV!
Combining frontier models from @claudeai with frontier agents at @Replit helped us discover a lot before major model releases, and helped us keep improving the agent experience for users.
Trying to break models is fun.
Claude: Before we ship a new model, these teams try to break it.
They build with it, push it to its limits, and tell us where it falls short. What they find makes the final model better.

NEW: Aleph Prover has formalized OpenAI’s disproof of Paul Erdős’ planar unit problem.
We are releasing the formalization as open source so that other researchers can inspect, extend, and independently validate the result.
See it here: https://logicalintelligence.com/blog/aleph-prover-erdos-disproof-lean-4-formal-methods
OpenAI: Today, we share a breakthrough on the planar unit distance problem, a famous open question first posed by Paul Erdős in 1946.
For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids.
An OpenAI model has now disproved that

Loving the @latentspacepod breakdown of our Laguna M.1/XS.2 Technical Report! The Latent Space paper club just did a deep dive, and their takeaways perfectly capture what we set out to build with our Model Factory. A few quotes from the video 🧵👇 (1/6)
https://youtu.be/QLfZamyMls0

The best design work doesn't happen in a chat box. You need space to explore ideas, create variants, and iterate
Meet the new Replit Canvas
Your agentic design tool to build beautiful websites, apps, marketing assets and more

I designed and shipped my UI identification game using Canvas, will share a walkthrough soon!
Replit ⠕: The best design work doesn't happen in a chat box. You need space to explore ideas, create variants, and iterate
Meet the new Replit Canvas
Your agentic design tool to build beautiful websites, apps, marketing assets and more

we're funding a lot of PhD researchers at YC these days
super pumped that we're hosting a reading group at the og YC office near Stanford and will be putting the videos on youtube. More of this to come!
Y Combinator: Last week we hosted the first ever YC Paper Club in Mountain View.
We brought together great AI researchers and founders to discuss both the state of the art and what it actually takes to get it into production.
Thanks to the following presenters:
0:12 - Intro from YC Visiting

The @SFDemocrats, @uniteddemclubsf, @SFENDC, @EdLeeDems, @AliceLGBTQDems, @sf_cadc, and @D2Dems all agree - No on Prop D.
We shouldn’t setback San Francisco’s comeback.
Vote NO on Prop D.

She bartended through law school. Spent 15 years prosecuting violent crime, elder abuse, and public corruption in SF. Now she's running for judge.
"Bar to Bar to Bar" by GL Studios. Vote Phoebe Maffei for Judge. June 2nd.

If you're building your own cloud agent like Devin or Ramp Inspect, there's lots of great details here on setting up VMs, computer use, memory, and more. Fun deep dive with the creator of OpenInspect on what setting up a cloud agent entails
https://www.latent.space/p/cognition

something like 99%+ of economists agree that building market rate housing reduces rents. Connie Chan's views on housing are equivalent to believing the world is flat
vote @Scott_Wiener

Canvas is my favorite medium to express my creativity.
Generate videos, images, SVGs... brainstorm with your team, add your favorite spotify playlist for background music, go deep in research, make silly websites!!! The world is your oyster...
UNDER DA SEA 🐟 (inspired by old windows OS fish games)...
Replit ⠕: The best design work doesn't happen in a chat box. You need space to explore ideas, create variants, and iterate
Meet the new Replit Canvas
Your agentic design tool to build beautiful websites, apps, marketing assets and more

I interview dozens/hundreds of new grads, nearly every day of the year. These are people with a well-formatted resume and a Bachelor's Degree in Engineering from well-regarded US universities and a GPA above 3.6. The majority cannot engineer, cannot function independently, cannot answer basic technical questions. We have watered down standards and inflated grades to the point that a bright, enthusiastic student spending four years in school sends almost no signal at all.
What does? Hard evidence of actually building stuff. There is no substitute for actually doing the thing.
San Francisco Chronicle: More than five years after the UC system lifted its standardized testing requirement, a coalition led by UC Berkeley math professors argues the drop in students’ math levels is “severe.” https://www.sfchronicle.com/california/article/uc-faculty-math-science-testing-22279842.php?taid=6a17d08a86566d0001e05981&utm_campaign=trueanthem%2B3988&utm_medium=social&utm_source=twitter