Principal Architect at @CopilotKit. Building the frontend stack for agents. Creator of Hashbrown, co-creator of NgRx, Google Developer Expert in Angular.
Hashbrown is a framework for building tab-to-complete in forms, predictions & suggestions, and AI chat powered by your service layer and components. It even lets you run LLM-generated JavaScript right in the browser.
Build generative user interfaces in Angular and React with https://t.co/OgjzACkwdC
Everyone has an opinion on AI in frontend dev.
We’re bringing together @ErikCH, Saurabh Rob Dahal, @DThompsonDev, @shrutikapoor08, @MikeRyanDev & @bytesofbree to discuss what what is actually working with AI ⚡
🎟️ Secure your spot: https://t.co/yAcPRuvXVW
We're at @CascadiaJS and the booth is BUZZING 🌲⚡
We're set up and ready — swing by the CopilotKit booth for:
🪁 Live AG-UI + Generative UI demos
✨ Dope CopilotKit and AG-UI stickers
👕 Fresh merch (the purple tees are going fast)
Come by our booth to see demos of production grade Agentic UI,
Build Agents that render UI components live inside your app!
Come say hi 👋
To allow agents to generate native UI beyond just text, we are building A2UI compose render.
We caught up with the team from CopilotKit to talk about bringing A2UI to mobile. Check out the code → https://t.co/qC8fXZf2I1
Make your Agent speak your UI
Many open source SDKs give you a simple chat UI
But only @CopilotKit lets you bring fully-immersive Claude and Cursor-like agentic experiences into your own apps.
Here's a thread on how to do, it in one sitting 🧵
We've raised $27M to build @CopilotKit — the Agentic Frontend Stack connecting humans & agents.
Because all UI will be AI.
Co-led by Glilot Capital, NfX and SignalFire.
We've raised $27M to build @CopilotKit — the Agentic Frontend Stack connecting humans & agents.
Because all UI will be AI.
Co-led by Glilot Capital, NfX and SignalFire.
Redux was always the right blend of pain-for-gain for me. The pain was the amount of code you had to write by hand and the complexity it entailed. The gain, however, was a level of predictability and safety that I felt comfortable with.
Agentic coding is pushing me toward better architectural choices because I can manufacture code on demand and use conversations to quickly rebuild my mental model of the codebase.
For example, @DavidKPiano's XState is just a joy to use with agents. I can generate and validate state machines in natural language, then use them to guide and validate agents' plans.
Similarly, leaning into @CoqLang to write validation engines for protocols. Why not have a body of theorems you can prove to validate critical contracts?
Articulating ideas with these tools wasn't the hard part: it was my slow transfer speed from idea to implementation.
A lot of teams have moved past trying AI and are now trying to make it hold up in production.
What actually works once #AI is part of real systems? What still feels unstable? What do teams have to change to make it reliable across the full workflow?
On May 7, @ElliottFouts, @mikeryandev, and @dkundel are talking through what they’re seeing firsthand and how teams are adapting as this moves from experimentation into day-to-day use!
📅 May 7 · 12:30 PM ET
RSVP: https://t.co/kd0t1dWwxq
I've joined @CopilotKit as director of product to build generative UI! Ridiculously excited to get back to my frontend roots while staying at the forefront of agentic development.
Post linked below on why I'm betting big on generative UI 🔗
@AlemTuzlak@solid_js I'm in this tweet and I don't like it. 😂
@solid_js is legit though. Inspired @angular's signals implementation. You might dig it. 👀