David Chouinard

Yearly reminder to watch Feynman explain fire: youtube.com/watch?v=N1pIYI…

How Anthropic’s product team moves faster than anyone else I sat down with @_catwu, Head of Product for Claude Code at @AnthropicAI, to get a peek into their unprecedented shipping pace, how AI is changing the PM role, and how to be the right amount of AGI-pilled. We discuss: 🔸 How Anthropic’s shipping cadence went from months to weeks to days 🔸 The emerging skills PMs need to develop right now 🔸 Why you should build products that don't work yet—then wait for the model to catch up 🔸 Why a 95% automation isn't really an automation 🔸 Cat’s most underrated AI skill (introspection) 🔸 What Cat actually looks for when hiring PMs now (hint: it's not traditional PM skills) Listen now 👇 youtu.be/PplmzlgE0kg

Today we're launching a rebuilt version of Claude Code on desktop. The app has been redesigned for the ground up to make it easier than ever to parallelize work with Claude. I haven't opened an IDE or terminal in weeks. Excited for you all to give it a shot!

@mnedoszytko @shack15 @AnthropicAI Congrats!

Yesterday at @Shack15, the winners of @AnthropicAI hackathon "Built with Opus 4.6" got to demo live — in front of builders, VCs, and one of the most incredible communities I've seen in tech. They flew in from all over the world to attend the ceremony. I have scored 🏆 3rd place out of 13,000 submissions and had the privilege to live demo PostVisit.AI - AI based system that guides the patients after they leave the doctor's office. And what a ceremony it was - @Claude Code's 1st birthday party. Here's what stood out at the event: among the winners were a lawyer, a cardiologist and a public road inspector. This hackathon proved something important — with modern AI tools like Claude Code, people with deep domain expertise can finally build the projects they've been dreaming about for years. And we all built them in 7 days. Huge thanks to @AnthropicAI, @iporollo,@adocomplete , @mskaciej and @cerebralvalley for an unforgettable event. And nothing beats munching Belgian chocolates with @bcherny between the demos. The age of the builder-expert is here. Congratulations to the fellow winners and all the builders from the community we have created. @breezwoods @wblueu @kevincollinsirl @ghatageanup @DesoukiNour

@morganlinton Thanks for checking it out!

This look super cool, will be testing this weekend.

@jaykreps Thanks for checking it out!

Among the zoo of security tools every org ends up buying this is the one that you most wanted: Something that just reads the code and fixes the security problems.

@DerekCedarbaum @bcherny Yep! Please apply here and include the project name under the company/organization field: claude.com/contact-sales/…

@bcherny Any chance you’ll make this free for open source projects?

We've been working on this for a while -- it's impressive (and scary) to see the kinds of security issues it has identified. Rolling out slowly, starting as a research preview for Team and Enterprise customers.

@ericabrescia Thanks for taking a look!

The frontier labs are moving up the stack at a pace I've never seen before. Impressive and certainly makes our jobs interesting...

Very important work. Come join us!

@benitoz @claudeai @bcherny @trq212 Thanks for the support!

@claudeai Wow @bcherny you don’t stop. Nicely done @trq212

Introducing Claude Code Security, now in limited research preview. It scans codebases for vulnerabilities and suggests targeted software patches for human review, allowing teams to find and fix issues that traditional tools often miss. Learn more: anthropic.com/news/claude-co…

@RowanLLane Thanks for checking it out

This is game changing, while engineering with agents, sometimes security vulnerabilities slip through the cracks. Claude introduces Claude Code Security which eliminates really, the only flaw of agent engineering.

@solana_sailor @AnthropicAI @claudeai Thanks for checking it out

🔐 Claude Code Security: AI Starts Securing the Code Era @AnthropicAI just announced Claude Code Security, and despite sounding like a developer update, it signals something bigger: AI is beginning to secure the world it is rapidly helping create. Here’s why it matters. 1️⃣ AI broke the old security model Software development has changed overnight. AI generates code faster than humans can review it. Shipping speed increased. Audit capacity didn’t. > Result: more code, more vulnerabilities, more attack surface. Security became the new bottleneck. Claude Code Security exists to close that gap. 2️⃣ From coding assistant to security reviewer Claude is no longer just helping write code. It now: > analyzes entire codebases > detects vulnerabilities > reviews pull requests > flags risky patterns before deployment AI moves from creator to reviewer. Not just building software, but judging it. 3️⃣ The agent era needs defense systems AI agents can now execute commands, modify environments, and interact with real infrastructure. That introduces new risks: > prompt injection > unintended actions > hidden security flaws at scale If AI can act autonomously, it must also monitor itself. This launch is Anthropic preparing for that reality. 4️⃣ The emerging loop: AI securing AI We are entering a recursive system: AI writes code AI audits code AI suggests fixes Humans supervise Security becomes continuous instead of reactive. Software starts developing an immune system. 5️⃣ Why this matters long term Three structural shifts: > Security becomes AI-native rather than manual. > Speed and safety stop being opposites. > Infrastructure becomes partially self-regulating. The industry is quietly moving toward autonomous reliability. 💡 This isn’t just a cybersecurity feature. It’s a transition moment. AI is evolving from tool → collaborator → guardian. The most important technological shifts rarely look dramatic at first. The tide changes long before the waves do ⛵️

@ExampleTestcase :)

@DavidChouinard I know someone but you hired him already

We're growing the Claude Code team and I'm looking for someone obsessed with evals. Not just writing them — designing the right ones, QAing signal vs noise and building the infra to run them at scale. It's deeply meaningful work and honestly the most fun I've ever had professionally. If you're technically curious and high agency, DM me. job-boards.greenhouse.io/anthropic/jobs…

@EricBuess Have him apply!

@DavidChouinard Hope you find an awesome fit!

@Sohailmo Will read for sure!

@DavidChouinard lol hey david, sorry I wrote you an essay of a dm but hope you get the chance (despite how much you’re probably flooded) to read it tldr I work at Amazon as an L5 and do interpretability research in my free time

Launch day!

New Anthropic research: Measuring AI agent autonomy in practice. We analyzed millions of interactions across Claude Code and our API to understand how much autonomy people grant to agents, where they’re deployed, and what risks they may pose. Read more: anthropic.com/research/measu…

Exciting to hear Spotify CEO's support for Claude Code in their recent earnings call We're increasingly seeing devs explain their goals and constraints in a spec and then delegating the entire implementation to Claude Code

CEO of Shopify @tobi is shipping more code than ever. 2024: 94 commits 2025: 833 commits 2026: 957 commits (in first 45 days of the year) Claude is turning CEOs back to builder mode.

Love seeing how Spotify is shipping with Claude Code. Their best developers haven't written a single line of code since December, they fix bugs from their phones, and they shipped 50+ features from Slack during morning commutes techcrunch.com/2026/02/12/spo…