Jess Yan

Can't wait to see everyone here! We'll be talking a lot about CMA 🙌🏻

@mc_anthropic You're describing yourself too 😉it's the best time ever to be a very brave (potentially crazy) generalist 🙌🏻🙌🏻

Jess is the one of the best product leaders I've ever worked with. She's constantly pushing the boundary of her role and revisiting what great looks like in the era of AI. Read this excellent blogpost.

@bedriyan0 Thank you for building with us 🙌🏻🙌🏻🙌🏻

Humbled and honored to win gold at the Built with Opus 4.7 hackathon. 🥇 It's an incredible feeling to win a hackathon with the tool I've used more than anything else this past year. Maybe that's exactly why. Deep gratitude to the judges who took such a close look at what we built: @lydiahallie, @_catwu, @adocomplete, @nosajab. And to @bcherny and @trq212 — thank you both as well; your work shapes everything we get to build, and is a constant inspiration. Thank you @claudeai, @claudedevs, and @cerebral_valley for the week, the platform, and the trust. 🙏 Medical education deserves better tools — and I couldn't be more excited to build them. #BuiltWithClaude #ClaudeCode #BuiltWithOpus

Another Claude Code hackathon comes to an end. Thank you to everyone who spent a week building with Opus 4.7, and to @cerebral_valley for co-hosting. Introducing the winners:

I recently had a chance to reflect on how the product role has evolved along the AI exponential. Here's a look into my experience building Claude Managed Agents, plus practical examples of the DIY'ed agents currently unlocking my productivity. claude.com/blog/product-d…

@zenoshin @claudeai Zeno -- the multiagent traces should be on! Can you DM me, might be something off with your gating

We had to build our own subagent traces + UI because coordinator agents delegating to sub-agents doesn’t surface the traces properly. Multi-agent delegation for @claudeai Managed Agents are cool… until you can’t see what your subs are actually doing 😂

Claude Managed Agents Memory is now in public beta! claude.com/blog/claude-ma… agents operate in isolated context by default - make sure learnings carry over by using our OOTB memory primitives.

@ntropiq @unstoppableweb you're in :) looking forward to your feedback!

@jess__yan just applied for access for @unstoppableweb, can you dm me?

Turns out the best way to handle lots of demand for Claude Managed Agents is actually... spinning up a few Claude Managed Agents. Keep the feedback coming!! 🌊

@pheuter @michael_chomsky @AnthropicAI Yes! We do plan on this. Likely will be built in rather than a toggle.

@jess__yan @michael_chomsky @AnthropicAI any chance u guys are planning to support something like programmatic tool calling or buffering outputs to a file when large for the managed agents MCP connector? this is the biggest blocker preventing me for replacing our custom Vercel Sandbox + Claude Agent SDK integration.

Anthropic managed agents is a work of art. It's solving a lot of problems nobody has even come close to solving. It's not just one or two innovations, it's like 20. Some serious thought went into this. Incredible work by @AnthropicAI. If you're building in the space, I'm implore you to take notes.

@vercel_dev Can't wait to keep partnering on this!!

Build a Claude Managed Agent on Vercel that: • connects to GitHub, Notion, Slack • securely handles user auth with Vaults • streams responses via Workflow SDK vercel.com/kb/guide/claud…

@bran_don_gell Thanks for building with us!!!👏👏👏

Thousands of people do their best writing with Spiral. And now you can do it directly from any agent.

@blockitAI Thanks for building with us early! We couldn't have done it without you. Please fix my calendar 😂

blockit 🫶🏽 anthropic claude.com/blog/claude-ma…

One of the best parts of building CMA was working with forward thinking, disruptive startups like @andocorporation to shape our features. Managed Agents first and foremost is an ecosystem. Thanks for building with us early, @saradu!

@effyyzhang You CRUSHED IT

Incredibly proud of my team, Claude Managed Agents is in public beta! If you've built a long-running agent you know the pain: state, infra, reworking your loop every model upgrade. This handles all of it. Go build something that runs for hours, not seconds.