SF City Attorney David Chiu says sending a confidential legal brief to reporters is grounds for removal from public office.
Jackie Fielder should resign.
Garry's List: After 3 months of medical leave, Supervisor Jackie Fielder returns to office next week.
Fielder's office has been under investigation since February, after a confidential City Attorney memo about legal risks related to a new sobering center was leaked to Mission Local.
A public
we are going to have to Rebuild So. Much. Infra. for the age of Software Factories
Zach Lloyd: http://x.com/i/article/2069756055639281664
Take one hour tonight to follow my Hermes tutorial below to install the app, set up your Telegram bot, enable voice, connect Google Workspace, and create a few routines.
I made it super beginner friendly and walked through all the steps live in 45 min. You won't regret it.
📌 Watch here: https://www.youtube.com/watch?v=2ZacwCbiLjg
Peter Yang: I love using Hermes from @NousResearch as my 24/7 AI chief of staff because it's reliable, free, and open source.
Here's my new 45-minute free course that covers how to:
→ Install Hermes safely
→ Connect Telegram, voice, Google Workspace, and more
→ Create routines like
Laude Institute
At @CAISconf last month, @andykonwinski sat down with researchers on the conference floor -- @matei_zaharia @istoica05 @lateinteraction @dawnsongtweets @gneubig @pgasawa @JonSaadFalcon @heathercmiller @ryanmart3n @alexgshaw @profjoeyg @swyx and Ioannis Ioannidis -- to talk open vs. closed, agent evals, compound AI systems, and where the frontier is headed 👀
https://www.youtube.com/watch?v=PbymvxpbWHg
Diana
most LLMs benchmarks score output, but not many measure the path to get there
be interesting to see more grading for reasoning traces
I told @samstphenson today that if I don't have @meetgranola on in a meeting I basically feel naked.
Perhaps the best product market fit signal lol
Peter Yang
Ok, I am interviewing the legend @trq212 on Friday. I'm planning to ask him to show me:
→ His Claude Code setup and how he uses /goal and /loop and dynamic workflows
→ How he does planning with HTML + artifacts
→ His best practices for Claude Tag
What else should we cover? Let us know in the replies.
Thariq: Claude Tag is an incredible new form factor for agents, so I think it's going to take some time to figure out the best practices, but these are some of my favorites 🧵
lots of folks prepping talks next week (congrats!). Some thoughts from RLing on thousands of hours of engineer- and researcher- focused talks:
- AI generated svgs > AI generated imgs. MAXIMUM 4 ai slop images in your slides, I don't care how pretty your mom thinks they are (exception ofc if your talk is ABOUT imagegen)
- Be pointy. Better to have 1 message with 5 surprising applications, than 5 messages with no concrete examples.
- Put code on screen. Engineers like to see code. Especially if they can nitpick it to death over things that aren't the point.
- Don't forget to Entertain. Being actually funny, or good at telling relevant anecdotes, is more important than adding yet another bullet point.
- Have a Thesis. every talk gets ONE "if you remember one thing from this talk, it is this" card. 1) SO MANY PEOPLE DON'T USE IT. 2) SO MANY PEOPLE DON'T PLAN FOR IT. you get one. use it well?
- Have a Thesis Slide. you've been in that talk where everyone gets their phone out to take a photo of the slide. Because people see 1000x more images than videos, you are much more likely to have a viral talk/slide if you have a single viral slide. You are much more likely to have a viral slide if you TRY and most people do not TRY. Struggling? Collect examples from talks you like and adapt their format to your thing. Your slides have a power law - don't spend 5% of your time each on 20 slides, spend 80% of your time on 1 slide and have the rest build up to that 1 slide.
- You might not need slides. live demo in IDE, off the cuff rambles, pull audience member up to roleplay, do call and response with the audience, sing/perform, voice over vibe videos, I have seen it all. higher risk/effort, higher reward when done well.
- Be pleasant to listen to. Talks with bad/no visuals still can be listened to. Talks with bad audio are DOA. Be confident, project, remember vocal variation, evoke emotion.
- Design the emotional journey. Start strong, end strong, have a peak aha/laugh/thesis moment in the middle. Everything else is buildup.
- Data driven talks are underrated. Present pretty charts and surprising, authoritative numbers. Use the stage authority to infer from the data to support the broader thesis. Done well, it will feel like my obvious conclusion from objectively looking at the data you presented, not you feeding the conclusion to me.
- How to shill your product/company without feeling salesy: teach me everything I didn't know I should to know about the problems you solve, and then you shall have earned the right to convince me you are the guys to trust to solve it once and for all.
- Actively watch a lot of talks. like with anything you both need a lot of reps AND you need to explore to find the styles/role models that you shine in. Passive = no introspection after watching, just mindlessly autoplay next video. Active = trying to articulate why a talk was good/bad after watching. If you want to REALLY hone it, think about how you feel about a famous talk and then TRANSCRIBE the talks and read the words on the printed page, and then compare with a "normal" talk and define rules you will follow for yourself to improve.
thanks to @dexhorthy for organizing the AIEWF Speaker prep meetup tonight. we should actually do more of these....
RT @frank_smith1111: In May SFUSD tried to secretly overhaul how public high school students are graded to promote so-called “EQUITY”… and…
Frank Smith
In May SFUSD tried to secretly overhaul how public high school students are graded to promote so-called “EQUITY”… and almost pulled it off.
No board vote. No parent notice. Hidden in a three-word footnote buried in a 25-page meeting agenda, timed to drop the last week of the spring semester while parents were distracted by the budget cuts / strikes.
The plan:
→ 80% grade = A
→ 21% grade = passing D
→ Homework? Doesn’t count
→ Late to class? Doesn’t count
→ Absent? Doesn’t count
→ Failed the final? Retake it as many times as you want
They’d already hired a consultant for $172K to train teachers over the summer. It was happening.
One school board member noticed. A local reporter published it. Within 24 hours, the mayor, Ro Khanna, and the internet exploded.
The superintendent’s response? “there was misinformation.”
The plan is paused, not cancelled. LA and San Diego already have versions of this running. SF will try again.
This is what happens when ideology runs a school district with no accountability.
We need to vote in competent board members and oust Maria Su.
Full story: https://thevoicesf.org/grading-for-equity-coming-to-san-francisco-high-schools-this-fall/
Claude Design is pretty great.
I gave it a repo for a mobile app I'm building and it reproduced the screens perfectly.
Except after one prompt it's telling me to save tokens already 😅
Is Widow's Bay worth watching, I'm not great with jump scares.
AI will bring forth an unprecedented surge in entrepreneurship.
From 'solopreneurs,' to the revitalization of the small & medium business segment, to the emergence of the largest companies of our times… providing the foundation of it all.
Max Gazor
Pause to consider who you listen to for advice.
https://ilyastrebulaev.substack.com/p/the-2026-venture-ranking-the-top-1f6
Dereliction of duty by media
🤝
Dereliction of duty by elected officials
💥Susan Dyer Reynolds🗞️: Joe Fitz has always been an operative for far left politicians. He needs to read the comments —readers see through him, too, finding this to be a dishonest “puff piece” and “in need of an editor.” I’ll never forget Fitz defending Preston when he tried to shut me up. Never.
SF lesson: You need to expose the extremist left in profound ways
The death of the innocent
The breakdown of public safety
The downfall of schools
These are all inevitable in the hands of the inept DSA
It will be the defining fight for the Dem party
Matthew Yglesias: Chicago’s left already staged a successful insurgency, but the incumbent left-wing mayor who led the successful insurgency is an unpopular failure so people don’t talk about him anymore.
Question is whether the reformist counterinsurgency will spread from SF to the Windy City.
Ruben Harris
1/ Today, @timurmeyster, @JohnHenryStyle, and I are launching The New Normal.
A new video show featuring the enterprise leaders actually deploying AI in production at scale across higher ed, insurance, and travel.
Operators behind the work. Not the vendors. Not the pundits.
http://thenewnormal.fm
Full YouTube Episode Here: https://youtu.be/r8ge3VbVFoA
Our best @aisdk release yet
AI SDK: AI SDK 7 is now available.
Introducing: reasoning control, agent-level tool approval, tool and runtime context, file and skill uploads, MCP Apps, durable workflows, terminal UI, sandbox support, harness integrations, telemetry, lifecycle events, and more.
Ayush S
Today, we're announcing a $60M Series B led by @BatteryVentures, bringing our total funding to $85M in just under a year.
Also joining the round are founders and operators who’ve built generational companies of the last two decades – @tobi (CEO, Shopify), @arashf (Dropbox), @chughesjohnson (Stripe), and more.
The round came together in 6 days. Here's why.
Every major category in enterprise software is seeing multiple AI-native challengers. CRM, ERP, ITSM – all being rebuilt from scratch by a new generation of companies applying AI to solve persistent problems we couldn’t before.
Employee Management (also known as HCM) is the exception. It’s the last frontier, and we believe the most important one. The operating layer to manage people, run payroll, benefits, compliance, and IT, for every company in the world, is still built on architecture that predates AI by decades.
This fundraise is the story of how Warp is changing that. The average Warp customer is growing 5x faster than their peers, with 1/10th of the HR and admin overhead.
We’re seeing a massive shift happening in how the best companies run their people operations. From the fastest-growing AI-startups to massive public companies, the winning teams are running lean: HR, finance, and ops generalists who automate as much as possible, and use their time instead for strategic work that AI can’t automate.
Warp is the platform of choice for ambitious companies operating at this new pace. Legacy HCMs help humans track the work. Warp uses AI to proactively complete the work.
Workday was built for the last era. We're building for the next one.
And it’s working. We've –
- Doubled ARR in Q1
- On track to $2B+ payroll volume this year
- Signed enterprise customers with thousands of employees
- Launched entire product lines back-to-back: Warp benefits brokerage and Warp Fabric (our AI-native IT automation suite built in-house).
A few thank-yous:
1. Our customers, the fastest-growing companies in the world, who trust us with their most critical systems. We wouldn't be here without you.
2. Our team - 50+ people in NYC who've built this platform, taken on the hardest problems in business-critical software. We're just getting started.
3. Our investors doubling down in this round, and some of our earliest believers – @sound_ventures_ (@aplusk, @epsteineffie), @PeakXV, (@arnavsahu341), @harjtaggar at @ycombinator, @balajis, @kevinhartz, @kvogt, @amasad, @HOFCapital (@myfady), @colinevans (@OpenAI)
We're here to arm ambitious American companies with Workday-grade power, but with the usability and delight of an Apple product.
With this new funding, we plan to fund deeper AI agents, tax and compliance infrastructure, expand our product suite, and support even closely our fast-growing customers.
Come join us.
Jarvis-core is the new most powerful launch video meme
Parker Conrad
We launched Rippling Data Cloud today - an all-in-one rebuild of the modern data stack, with AI deeply integrated throughout.
Why would you want an org-and-employee-centric data stack? Well, here’s how I used Rippling Data Cloud to help with token burn and cut AI slop. 1/
https://x.com/parkerconrad/status/2070169982827192525
Parker Conrad: http://x.com/i/article/2069974661891235840
Revant Teotia
🎉 Our paper MJEPA has been accepted at #ECCV2026 !!
Huge thanks to my awesome collaborators @AdrienBardes, @michaelrabbat , Sumit Chopra @mattmucklm and Nicolas Ballas
Paper: https://arxiv.org/abs/2606.25225
Code + checkpoints coming soon
See you at ECCV!
Legora is the defining legal AI startup
Max Junestrand: Legora went through @ycombinator in W24. I know firsthand how hard it is to navigate legal as an early-stage founder.
Today, @CooleyLLP is launching Cooley GO Lab using the @WeAreLegora Portal, bringing their knowledge and workflows directly to YC founders, so legal needs can
Ankit Gupta
I love that in this arms race for speed there is a seed stage YC company @wafer_ai fighting for the top spot
Yuchen Jin: You may have heard that GLM-5.2 at 328 token/s is cool,
How about 392?
Databricks is now #1 in inference speed for GLM-5.2 on Artificial Analysis. It's a great model, and we did a lot of optimizations.
Matt MacInnis
This is the biggest product payload of my career. Data Cloud morphed from a BI platform into an insane AI data machine over the last 2.5 years.
Parker Conrad: We launched Rippling Data Cloud today - an all-in-one rebuild of the modern data stack, with AI deeply integrated throughout.
Why would you want an org-and-employee-centric data stack? Well, here’s how I used Rippling Data Cloud to help with token burn and cut AI slop. 1/
Paul Graham
Ten years in, Rippling is still shipping big new things.
If you're an enterprise customer, this is the safest kind of company to buy from. They're energetic enough that you know you'll get the latest tech, but you also know they won't just disappear.
Parker Conrad: We launched Rippling Data Cloud today - an all-in-one rebuild of the modern data stack, with AI deeply integrated throughout.
Why would you want an org-and-employee-centric data stack? Well, here’s how I used Rippling Data Cloud to help with token burn and cut AI slop. 1/
Something that I love about working at @vercel is just how enjoyable it is to use our technology. From using the platform to ship new products, to the agents you find in our workspace that make life easier. It's like experiencing the future while building the future 🔁
"Can @v0 use my design system?" you've asked.
Now the answer is: @v0 can use the exact same components your product uses in prod.
We put it to the test with @microsoft Fluent, @shopify Polaris, @ibm Carbon, @palantirtech Blueprint, @vercel Geist (ofc), and more. Watch ↓
v0: v0 Design Systems 2.0 is here.
Import your design system from GitHub, npm, Storybook, Figma, and more.
Build with your real components, colors, fonts, and patterns.
How do I reach an actual human at Google AdSense - is there a customer support team? They keep directing me to user forums 😭
Garry's List
In 2002, SF created a committee to advise on newspaper boxes.
Print circulation has since cratered ~90%+. The boxes are gone, but the committee still meets on an "as needed" basis—one of SF's many defunct bodies that continue to operate as the city runs a ~$642M two-year deficit.
Major updates to mobile app coming soon!
James: Replit is so satisfying to use
Agents are being adopted very quickly and accelerating work. How this looks across OpenAI itself:
OpenAI: Work at OpenAI is being transformed by agents, in every department.
Across our entire company, people are using Codex to do work that is more complex, longer-running, and increasingly cross-functional.
Our internal usage offers an early look at how agentic tools may reshape
One thing that still annoys me about even the best AI models and agents:
They’ll give an overconfident opinion. Then if you push back even a little, they’ll usually reverse it right away.
It makes me feel like they weren’t thinking that hard in the first place.
A good human teammate might change their mind too, but they’ll at least explain why or have a more principled approach to their thinking.
Maybe there’s a way to fix this with a system prompt?
Peter Yang
One thing that still annoys me about even the best AI models and agents:
They’ll give an overconfident opinion. Then if you push back even a little, they’ll usually reverse it right away.
It makes me feel like they weren’t thinking that hard in the first place.
A good human teammate might change their mind too, but they’ll at least explain why or have a more principled approach to their thinking.
Maybe there’s a way to fix this with a system prompt?
Atul Gawande
Independent analyses estimate that your actions to dismantle USAID and drastically reduce lifesaving foreign aid have already killed 700,000 people. Here are some of those people. 🧵https://x.com/elonmusk/status/2069445014787129590?s=20
Elon Musk: All DOGE required was contact information of the recipients to confirm that funding was not fraudulent. No validated medical funding was stopped.
Anything that appeared to be legitimate lifesaving funding continued and is now administered by the State Department.
If anyone had
Replit ⠕
Replit now supports 450+ integrations.
No matter what you're building, Replit Agent can wire it up to the tools you already use: payments, messaging, data, CRMs, design tools, analytics, and hundreds more.
Just describe what you want. We'll handle the connection.
Peter Yang
I am excited to attend @aiDotEngineer on Tuesday for the first time and the scale of this conference that @swyx and team put together is simply massive.
I am excited to attend @aiDotEngineer on Tuesday for the first time and the scale of this conference that @swyx and team put together is simply massive.
Mark Cuban
It’s time for everyone to realize that the fight against data centers has nothing to do with data centers.
They have become a proxy for the hate towards AI and the concentration and accumulation of wealth it’s creating.
Until those running the big LLMs understand this and start a community tour, not to explain the benefits of AI, it’s too late for that, but to help towns and cities that may be impacted by job losses (and I’m a believer their will be a net gains in a few years), this battle is only going to get more intense and let me tell you now , no matter how much money you pay to buy politicians and races, you will lose.
One thing I have learned is being hated is not good for business.
How can they help ? They will tell you. You will need to do what they ask. Billions of dollars is a lot of money across towns and city programs. Across the major LLMs, it’s a cost of doing business.
At the same time, I would go to LA and NYC and ask the arts and creative unions what kind of programs would help and protect their artists. DO NOT GO TO THE MUSIC OR FILM COMPANIES. that will make it worse.
Don’t try to pay famous people to endorse what you are doing. That’s dumb.
Talk to artists and ask them what you can do to provide financial and creative support. Every creative I know is TERRIFIED about what AI will do to their profession. You must meet them face to face and basically do what they say.
The big LLMs have lost the PR battle. Why ? Because they all suck at putting people first. They have an SV attitude that makes them all think they are John Galt saving the
world
Given the number of data centers and power that is needed, today and going forward , If you don’t kiss the asses of the people that go to work every day, and are just trying to pay their bills, you will fall far far short of the capacity you need to make your business work.
PROPAGANDA
Can you build an app before your next stop?
This Saturday, June 27th. New York City.
One train ride. Build a product using @replit by the end of the line and win!
RSVP ⬇️
Grok Imagine Video has become the clear leader in @vercel AI Gateway, now amounting to ~50% of videos generated by developers 🤯
http://vercel.com/ai-gateway/leaderboards/video/models
Dan Gray
Re @contextconor “There’s a weird phenomenon among VCs where the less successful they are, the more evil they become to founders to squeeze more money out of their best startups out of necessity which then becomes a vicious cycle of adverse selection.”
- @garrytan
https://credistick.com/adverse-selection-in-venture-capital/
Why even visit your email inbox anymore.
btw as a musician myself i've gone ahead and pulled the trigger for the first ever music corner at AIE
if you code + music, we want impromptu jams between sessions. grab me to sing with u if u want. give us your best dad rock.
sponsorship for this + networking night on jun 30 are the only show floor opportunities left at aie, everything else is sold out
(btw tix on track to sell out TOMORROW)
AI Engineer: 6️⃣ Things to Know about
AI Engineer World's Fair 2026
- It’s bigger than all previous AIEs
- 4x Larger Expo with 4 Expo stages
- Researchers: Poster sessions & Poaster sessions
- AI Leadership: Token Billionaires & Off the Record
- AI Verticals: Healthcare, GTM, FDE, AGC,
Thariq
I'll be talking more about Claude Tag with @petergyang and at AIE with @_catwu.
Let me know if there's anything you'd like us to dive into more!
ClaudeDevs: Claude Tag is the next evolution of agents.
It's a proactive, multiplayer agent with memory and identity, built on top of Claude Code.
Learn more about how Claude Tag works and best practices for using it in this deep dive.
How we imbue coding agents with our design standards
Vercel Developers: To ensure coding agents live up to our design standards, we built a system with a skill, linters, evals, and an updating loop.
Learn how we did it and how to build your own ↓
https://vercel.com/blog/teaching-agents-product-design-at-vercel
Next.js "Ways to fix this" with "Copy prompt" buttons is a work of (agentic) art
AI Engineer
Stripe + Metronome are hosting a private after-party for founders & product leaders.
July 1, 6 – 9pm, SPIN, near Moscone West
Ping-pong and World Cup, plus founders and builders worth meeting.
Spots are limited & individually approved. RSVP:
https://luma.com/uehktron?utm_source=AIEWF_newsletter
JohnPhamous
an improving design loop
https://vercel.com/blog/teaching-agents-product-design-at-vercel
Latent.Space
In this episode, @OpenAI Chief Research Officer @markchen90 joins @allenpark to flambé shrimp, cook Korean stew, and chat about being at the frontier of AI research: why scaling laws and pre-training still matter, how OpenAI chooses research bets and allocates compute, what it means to develop research taste, why evals are in crisis, how to avoid benchmark-maxing, and what it will take for models to handle long-horizon real-world work, multimodal reasoning, and eventually end-to-end AI research.
Timestamps:
0:00 Intro
0:28 The Soup Story
1:52 From Trading to AI Research
3:21 How to Develop Research Taste
5:23 RL, Evals, and Superhuman Benchmarks
8:17 Cooking Begins on the Impulse Stove
8:53 Scaling Laws, Pre-Training, and Reasoning
12:33 OpenAI’s Research Roadmap and Compute Allocation
15:48 What Makes a Great Researcher
19:33 The Evals Crisis and Benchmark-Maxing
24:34 Jagged Intelligence, Context, and Long-Horizon Learning
27:14 Shrimp Flambé and New Research Bets
31:32 Multimodal Models and One Architecture
32:36 Vibe Researching and End-to-End AI Research
34:36 Failed Bets, Postmortems, and OpenAI’s Alpha
37:07 Final Taste Test
37:53 Overrated vs. Underrated AI Research
41:00 Closing
AI Engineer
The first WF2026 keynotes are premiering now!
Congratulations to:
- @techgirl1908, Agentic AI Foundation:
Build Systems, Not Code
- Nishant Gupta, Meta Superintelligence Labs:
Production Evals For Agentic AI Systems
- @pauliusztin_, Decoding AI:
Turn 10,994 Notes Into Your Agents' Memory
- @victorsavkin, Nx:
A Genius With Amnesia
- Isadora Martin-Dye, Isadora & Co.:
Stop Writing Tone Instructions. Layer Them.
- @raw_works, OpenProse:
Recursive Coding Agents
- @ishaansehgal, Omnara:
The Log Is The Agent
- Jacob E. Thomas, Results Generation:
The Miranda Hypothesis: How Hamilton (the Musical) Poisoned Your Persona Evals
- @jamesjellow, OpenGov:
Agents in Production: How OpenGov Built and Scaled OG Assist
the rest of our online track drops this weekend, get hyped!
Angie Jones: While building agents, I find myself back in my dev flow state.
In my AIE online keynote, I'll walk you through designing an agent and highlight all the core engineering principles that pop up along the way.
Live today at 2p ET
digital ocean for running a codex remote session:
OpenAI Developers: You can now spin up a persistent cloud development environment in one prompt with the @DigitalOcean plugin for Codex.
It runs in your DigitalOcean account and keeps working when you step away.
😭😭😭 Who gets to have it early?
Amir Efrati: NEW:
Trump admin asks OpenAI to stagger GPT-5.6 release over cyber concerns. Will approve "access customer by customer during this preview period”
What a time we are in...
