Henry Mao
@Calclavia
Co-founder/CEO @ http://smithery.ai (@SmitheryDotAi) - orchestrating MCPs for AI agents, backed by @southpkcommons; Prev. co-founded http://jenni.ai ($10M ARR, exited)
The future of the internet will be dominated by tool calls, not clicks. We're building Smithery to orchestrate this new era of AI-native services for AI agents. Read our mission 👇
i just built an auto-triage agent using Github MCP with @SmitheryDotAI imagine the use cases for technical PMs! the agent can: - create issues - classify them with AI (bug/feature/docs/question) - auto-apply labels with confidence scores 🏷️ - post helpful triage comments…
ChatGPT Agent is so sick But you can only use it with microsoft office and like five other apps... and you can only use OpenAI models 😭 I built my own version using @ExaAILabs, @browserbasehq, and @NotionHQ MCPs in Claude Code. Check it out! Powered by @SmitheryDotAI
I just love Sonnet 4's enthusiasm after it spent 50+ attempts iterating the code and finally fixing all unit tests.
@SmitheryDotAI: the leading ai-native service orchestration platform! just finished reading the docs using claude! all I had to do was add `/llms.txt` to the URL like this: smithery.ai/docs/llms.txt
Proud to be the official host of the @browserbasehq MCP! Try it now: smithery.link/browserbase
Introducing the Browserbase MCP🛠️ We've updated our tools to be powered by @Stagehanddev, our AI browser automation SDK that is more powerful that direct Playwright calls. You can now use our remote hosted server and run browser sessions in parallel, unlocking more use cases…
APIs & data models are the artifacts derived from a clear thesis. Building one is a forcing function to codify your thesis in its most precise form. It's less about the API but more about having a sharp understanding of your users & product.
Patrick Collison: API Design Shapes Your Business Strategy "If I was to do everything at Stripe again, the thing we could have done differently would be to have spent even more time on APIs and data models."
ARC AGI 3 with games as a benchmark feels full circle. We started with agents playing OpenAI Gym CartPole and other games, now we're back to games!
Today, we're announcing a preview of ARC-AGI-3, the Interactive Reasoning Benchmark with the widest gap between easy for humans and hard for AI We’re releasing: * 3 games (environments) * $10K agent contest * AI agents API Starting scores - Frontier AI: 0%, Humans: 100%
Flight booking AI demos smell like the classic "solution looking for a problem"
Visited a startup's office @menloresearch where they might eventually build an MCP so LLMs can prompt their robot!
105,000 tool calls were processed for 20,000 users on @SmitheryDotAI just yesterday. If you want to build the platform to discover and distribute intelligence itself, we’re hiring. I’ve been a founding engineer before, and we’re building the company I would love to work for.
Do you use @todoist to organize your todo lists? I just organized my inbox and set due dates for all the tasks. All in one prompt using the official Todoist MCP server on @SmitheryDotAI!
Come say hi if you're in SF this weekend! We're co-sponsoring a hackathon Win a new Nintendo Switch or Polaroid Camera! Register: smithery.link/ff-hack
Smithery is all you need. Thanks so much @schiffyalex!
🛍️ Your agent can now discover and recommend products 💸 You get paid when it drives sales Shout-out to @charliebholtz and @kamathematic for making MCP so easy Build an AI stylist/co-pilot/storefront with @try_channel3 docs.trychannel3.com/mcp-overview
If you're truly building at the frontier, you should expect your priors to be wrong with reasonable probability. Conversely, if the outcomes of your bets stop surprising you, it means you should push harder and make bolder bets.
👀 I built my own minimal, functional MCP client: 👉 Next.js + AI SDK ONLY 👉 Hosted on Vercel with local storage for chat/config 👉 100% Open Source Check it out: dotcom.chat
openai spent $15m+ on "chat dot com" i just spent $6 on "dot com dot chat"