Meet AI is a video conferencing SaaS where AI agents join live meetings as active participants and respond in real time. I built the full application stack.

The hard part is the real-time layer. The backend provisions an AI agent the moment a meeting starts and keeps it running for the full call. Speech runs through a transcript pipeline and generates responses through the OpenAI API in near real time, so the agent replies while the conversation is still moving, not minutes later. After the meeting, the backend writes a structured summary through the same API, handled async by Inngest so nothing blocks the live app.

The stack is Next.js 15 and React 19 with the Stream Video and Stream Chat SDKs, tRPC and Drizzle ORM on Neon PostgreSQL, and Better Auth across the stack.

The platform runs 30 simultaneous live sessions, each with its own active AI agent, with no backend slowdown.

What the client gets: live AI agents that take part in real meetings, plus automatic summaries, on a stack built to hold many sessions at once.

Perfect

Built a voice-first AI app in @usewonder.

The goal was simple: create something that makes people instantly think, "Damn, this feels like a real AI product."

I started with a few foundation screens, imported them into Wonder, then used Wonder Chat to expand the experience, refine the UX, and iterate much faster than I normally would.

My favorite feature has to be Shaders. It instantly made the interface feel more alive and premium. I also experimented with MCP + Codex to push the prototype further. Still learning that workflow, but it was fun to explore.

Here's a quick walkthrough of the project and how Wonder fit into my design process.

P.S. Ignore my grammar and occasional stuttering in the video. My brain renders thoughts faster than my mouth ships them. 😭

🔗 Wonder file: https://app.wonder.so/breeje-anadkat/files/019f2310-4ec0-7ab2-9ef1-0d3ff83a5f4f