1/4 🚀 Catnip is live on the App Store! It’s my first iOS app — and it lets you run Claude Code *everywhere*. Same agent. Real dev environment. In your pocket. 📱

At @PWVentures, we've been quietly building AgentCribs, a private community of founders and hackers building custom agentic coding/business/marketing/etc setups and sharing wins and losses. Now, we're opening up the community a bit with our first event on May 6 in San Francisco! I'll be sitting down for a fireside chat with Peter Levine (legendary @a16z investor who sat on our board at @GitHub for many years) to discuss the bleeding edge of AI ideas and realtime insights on where the market is headed. The evening will also include an update from the core AgentCribs team, demos and discussion from people building with agents, and an invitation to participate in ongoing collaboration. If you've been hacking on your own real-world agentic workflows, built your own harness, 10x-d Claude, or just want to know what the future looks like, you belong!

I got a little carried away asking questions about the electromagnetic spectrum and had Claude build me an interactive tool to help me wrap my mind around it 🤯 vanpelt.github.io/sparky/too…

3/4 Under the hood, Catnip spins up a GitHub Codespace and connects your phone to it securely. You get a full terminal when you want it — or a clean native iOS UI when you don’t. GitHub gives personal users 120 free hrs/mth, and we automatically shut down inactive instances.

2/4 I built Catnip because Claude Code is 🔥 and it really shines when it has a rich, persistent dev environment. With Catnip, your agent lives inside your *actual* dev environment — plugins, tools, and all.

This is SOO cool. I’ve dreamt of this since we started the company. Love seeing this new era of TUI’s 🤓

Claude Code has been such a joy to use. I couldn't resist making a new tool. Introducing Catnip: github.com/wandb/catnip Catnip runs Claude in a container and makes running multiple agents a breeze. I'm ready for people to try it so I can use Catnip to improve Catnip 🤓

Our engineers have been elbows-deep with @wandb, and we cannot wait to show you what we’ve built together on day 2! 🫡

Our very own @GoldbergChen will be the CoreWeave keynote speaker at the AI Pioneer Speaker Series at @wandb Fully Connected 2025! Reserve your spot to learn from the innovators shaping what’s next for AI development: hubs.la/Q03rwLrj0

Haven't posted in a hot second, we've been busy. Can't wait to share some of the new things we're working on at Fully Connected in SF June 17-18th!

Our SWE-Bench submission has been accepted and is officially SOTA! Thanks SWE-Bench team for making such an important benchmark.

My o1-based AI programming agent is now state of the art on SWE-Bench Verified! It resolves 64.6% of issues. This is the first fully o1-driven agent we know of. And we learned a ton building it.

👻 This Halloween I dressed up as a Mad Scientist. My latest invention: Fester - a moving, dancing, speaking Skeleton that reacts in real time to kids costumes with a custom spoken greeting in a spooky voice. Presenting: Full tutorial (1st comment) reactions vid /1

Just heard a reminder to check your carryon for a firearm before going through security by my gate at BNA. Who was that message for? There are enough people flying that didn’t bring their firearm this time, but might forget next time?!

Hack-a-thon at our new office in SF this weekend: lu.ma/judge. Can't wait to see the demos!

I've been coding with my own command-line agent every day for the last year, and now you can too: `pip install programmer` 🧵

We're doing a meetup in our SF office on Thursday hosted by @sh_reya on scaling up GenAI model evaluation - register here: wandb.ai/site/resources/even…

Announcing new visual evaluation comparisons in W&B Weave. Easily compare model performance, latency, and token usage in a comprehensive view. Watch a lead engineer from Weights & Biases demo the feature: youtube.com/watch?v=oBVzAuYi…

We are excited to announce our new integration with @MistralAI. Fine-tune Mistral models with Weights & Biases for seamless experiment tracking and metric comparison. Auto-logging in Weave captures inferences, tool use, and more. Check out the docs: github.com/mistralai/mistral…