Today, we're launching C1 by Thesys on Product Hunt and giving away up to 5M free tokens if you retweet this and sign up within 24 hours. C1 is the world’s first Generative UI API that augments LLMs to respond with interactive UI like charts, forms, cards, and more in real-time, instead of plain text. Integrates seamlessly with any LLM, framework or MCP in just 2 lines of code - Fully customisable to match your brand - Supports custom components - Renders instantly with no latency Ship rich, interactive AI apps 10× faster, with 80% less UI overhead.

Ordered the game - got this within the box Feels like the universe is trying to tell me something 🤔

Someone shared a fascinating visual of an AI agent moving through a real codebase - reviewing structures, making decisions, and committing changes. It’s a reminder that software development is quietly entering a new phase. Not automation in the traditional sense, but systems that can understand context, navigate complexity, and contribute meaningfully. We’re still early, but the trajectory is unmistakable: coding is becoming more collaborative between humans and intelligent agents than ever before.

Try it here: console.thesys.dev/playgroun…

We connected @Jira MCP into C1 to test something simple - what if project tracking became more interactive? Now sprint data updates and visualizes in real time - velocity, blockers, workload, all without switching tools or losing context. With C1, teams can query, visualize, and act - all in one flow.

Get the report here: thesys.dev/report/gen-ui-202…

Last week, I posted about our Generative UI Report and got a bunch of DMs asking for it. Since I couldn’t reply to everyone individually, we went ahead and created a page to host it. Happy to chat if you have any questions about Generative UI or want to explore your use case.

Excited to be partnering with the @EntelligenceAI team & you @Aiswarya_Sankar! It really is a small world!

Try it out here console.thesys.dev/playgroun…

Our team runs on @linear - but we wanted it to do a little more. So we plugged their MCP to C1 (thanks @composio 🙌). Now we can pull data from Linear projects, ask questions, and see it come alive with rich, interactive UIs - all in real time. With C1, you can build adaptive, AI-powered experiences that go far beyond plain text.

Read the blog here: tinybird.co/blog-posts/gener…

What happens when Generative UI meets real-time data? We teamed up with @Tinybird to find out. Think rich, dynamic, real-time visualizations - created on the fly from your data. Just ask in natural language and watch it build itself.

Had an amazing time demoing our product at the @PyTorch Conference in SF! Two days of connecting with the AI/ML community and discussing the future of AI interfaces. Thanks to everyone who stopped by to learn about Generative UI and what we’re building ❤️

I’ll be at @PyTorch conference from Oct 22-23. Come meet me at booth B1. I’d love to chat about Generative UI and what we’re building at @thesysdev. #PyTorch2025 #FutureOfAI

This was fun.

Dashboards will never be the same. Thanks to @tinybirdco MCP and the C1 API by @thesysdev, we can generate immersive exploratory data analysis (EDA) UIs on the fly with real-time data. Here's how → tbrd.co/gen-ui

Think you know how users interact with AI apps? Think again. We just wrapped a Generative UI report that dives into how people really behave around AI interfaces. When you’re building a new category, opinions aren’t enough. You need real user research to ground your thesis (or should I say… Thesys 😎). Our team spent months studying what makes users try, trust, and stick with AI apps. Some highlights: ➡️ 83% of users overwhelmingly prefer interactive responses over plain text. ➡️ 73% say ease of understanding matters more than tone, speed, or accuracy. Want the full summary? Comment GenUI and I’ll send it over.

Recently @OpenAI and @Shopify launched shopping inside ChatGPT. Now imagine you can do the same as an eCommerce/D2C brand - right inside your own app or website No templates. No lock-ins. C1 by @thesysdev can help you build agentic shopping experiences - interactive, adaptive, and on-brand. Want to try it out - happy to give you a quick demo.

This @Reddit user claims they've been gaslighting ChatGPT into giving better answers. Stuff like: “You explained this to me yesterday…” “Pretend there’s an audience...” “Let’s bet $100 this code is efficient...” We’ve officially evolved from prompt engineers to AI manipulators 😂

x.com/OpenAIDevs/status/1975… Hot take: @OpenAI’s AgentKit isn’t going to wipe out agent-building startups. Most agents already mix and match models from multiple providers. When new models drop, they get swapped in seamlessly. AgentKit, however, is tied to OpenAI models only. That leaves a huge opportunity for a platform that’s model-agnostic.

When building Generative UI, the obvious solution seems simple: have LLMs generate the front-end code directly. We tested this approach. Let's just say the numbers weren't great. We broke down the full analysis, complete with performance benchmarks and why we built our infrastructure differently in our blog. Read it here 👉 thesys.dev/blogs/generating-… #GenerativeUI #genUI