A big thank you to everyone who joined us for the “AI for Engineering Leaders Dinner” last Saturday in Bengaluru! One of the highlights of the evening was seeing Ask Ellie in action. It also turned into a great space for hearing different perspectives on how AI fits into

MCP Apps feels like a new primitive for agents. We built Thesys MCP so Generative UI becomes a first-class citizen in agent workflows, not an afterthought. When your thinking happens inside Claude, it shouldn’t stay trapped in text. Humans reason spatially, visually,

AI assistants like Claude still reply with walls of text. Slow. Passive. Meh. What if they could show instead of tell? With C1 MCP by Thesys, Claude can respond with Interactive UI like 📊 Charts 📦 Cards 📝 Forms 📄 Slides If you use Claude, you’ll want this. Takes 2

Any good implementations of /compact in the wild? Didn't realise this fundamental block of conversational apps is such an overlooked problem

🚀 Agent Builder by Thesys just dropped! You can now build agents like Claude Cowork, Perplexity, Julius, Gamma — but customized for your own use case. What makes it different: - No rigid workflows or coding required - Responds with interactive UI (charts, tables, reports,

Today we take one more step towards making Generative UI accessible to everyone beyond just developers With Agent Builder by Thesys, building engaging agents is as simple as dropping your Excel files and asking away. The LLM plans, reasons, and adapts while you scan,

We built a better alternative to json-render & A2UI Up to 3x faster rendering & 67% fewer tokens. Far fewer broken outputs. We've been shipping Generative UI interfaces to 10,000+ developers for the past year. We know where JSON-based rendering breaks in production. So we built

We just published our benchmarks and source code: github.com/thesysdev/open… Check it out and let us know what you think!

watch openai acquire them in like 6mo

Rabi Shanker Guha and Parikshit Deshmukh told me about their vision for generative UI back in early 2024 when the category barely existed... LLMs responding with live, interactive charts and components instead of just text. They were years ahead on this. Cool to see the whole market

We deleted entire Rust codebase and replaced it with TypeScript and it got 3x faster. Rust is fast != Rust in browser is fast Here's what we found 🧵

I am very bullish on OpenUI (openui.com) .. it comes with great components out of the box, streaming protocol is super efficient and has a super small footprint! Also .. very easy to put together with LangChain JS 🚀

Had fun adding HeroUI v3 support in OpenUI - built a Generative UI version of Google Forms over the weekend (since HeroUI works across React and React Native this should just work as easily on mobile too)

Streaming-first language for model-generated UI github.com/thesysdev/open…

What if your agent could generate a beautifully designed personalized email for every user OpenUI + react-email by Resend

"A star goes supernova. Scatter wireframe debris outward." English → interactive 3D. Streamed. Generative UI isn't just forms and charts OpenUI + Three.js

notice something? Linear, PostHog, Attio - all shipped the same thing in the last few weeks. Homepage is a chat bar - not a dashboard. This is the SaaS industry quietly admitting that traditional UI doesn't work anymore. Every user is different. One homepage can't serve them

x.com/i/article/2040…

Replacing dashboards with a chatbar was step one Here's what step two could look like (Inspired by Basedash - they've been building in this space for a while)

1/ Got a flood of 'how to build that?' DMs on the dashboard demo Short answer: OpenUI Lang v0.5. It's the layer that lets the agent go beyond rendering UI and actually wire up a working app. State, Data & Actions - the three building blocks for real Generative UI