“Organizing the world’s information” is a solved problem but we barely scratched the surface of “organizing a person’s information”
What problem are you willing to work on for 7 years?
Andrew Wiles worked on Fermat's Last Theorem alone in his attic for 7 years and told nobody except his wife.
The problem was 358 years old. Fermat scribbled in a margin in 1637 that he had a proof and no room to write it, and every mathematician since had failed.
Wiles saw the statement in a library book at 10 years old. He never let it go.
He was a Princeton professor when he started in 1986. He stopped publishing new work and released old results in pieces to hide the silence.
He announced it in Cambridge in June 1993, over 3 lectures, revealing the target only at the end. Newspapers ran it worldwide.
Then a referee found a gap. One step in the proof did not hold, and the error was fundamental.
He spent 14 months trying to close it in public view, with the whole field watching. He was ready to quit.
The fix arrived in September 1994. He has said he stared at it in disbelief for 20 minutes.
The BBC filmed him describing that moment. He gets a few words out, then turns away from the camera in tears.
He kept the greatest result of the century secret so nobody could take it.
ESP32 back in the game @ilbert_luca
🚨 side project alert 🚨
Announcing Muse Gadgets, an open source ESP32 firmware and Linux sdk so that you can make hardware devices that work with Muse.
Grab an API token from gadgets.muse.ai and point your favorite coding agent at the github repo to build your own peripherals for Muse.
I feel like it's coming really really sooooooon 👀
Computers disappearing into the fabric of everyday life
Our biggest DevDay ever, don’t be late.
openai.com/live
Finally moved my personal context-use instance from AWS to nibrun.com
Bills went from almost 60$/month to 1$/month. context-use.com is also hosted there.
Reminder: you do not need a VM!
Thanks @ilbert_luca
Got the same feeling I had when I first started using the raspberry pi a decade ago... my computer can do anything now.
loving it @dhh
i love the shift from “integrate another saas” to “import a library”.
as agents do most of the plumbing, a lot of saas can be replaced by libraries + great skills.
getting rid of auth0 in favor of @better_auth was such a pleasure.
any other examples?
open-connector now bundles only the code and assets that a host actually needs: github.com/oomol-lab/open-co…
pretty useful when nibrun.com gives you just 256 MiB of ram
Open Connector will soon be headless. So you can run it within your project and bring your own dashboard: github.com/oomol-lab/open-co…
Soon on NPM: github.com/oomol-lab/open-co…
AI agents are only as powerful as the tools they can reliably access.
Nango built a solid foundation for open-source integrations. OpenConnector takes it further for the AI agent era.
OpenConnector open-source auth gateway for agents:
• 900+ providers + 9,000+ Actions with full inspectable schemas & contracts
• Self-host instantly with Docker, or deploy natively to Cloudflare Workers (edge performance, global scale, zero infra headaches)
• Production guardrails built-in: scopes, policies, redacted logs, connection identity
• Native interfaces: MCP, SDK, HTTP, OpenAPI + beautiful Web Console
• Your credentials stay in your runtime. Full transparency. No black boxes.
• Same clean contracts as OOMOL hosted (1,000+ providers) — start self-hosted, scale seamlessly when ready
No vendor lock-in. No fragile connectors. Agents that actually ship to production.
github.com/oomol-lab/open-co…
Quickstart:
docker compose up --build
If you're building real agent workflows, this changes the game. Star it and tell me what you're working on 👇
Every single app makes me pay for the intelligence I use. Why can't I just bring my own intelligence to the app?
I want a single subscription with one of the providers of this new utility and use it to power everything else. That's how data plans work.











