Full-Stack AI Developer
I build GenAI-powered full-stack products with a focus on real-time systems. Currently interning at Voiceshop.in, contributing to open-source, and deepening expertise in LLM integration patterns and scalable Next.js architectures.
Production-grade platform where custom AI agents join WebRTC meetings with real-time transcription, intelligent summarisation, and in-meeting tool calling.
- Custom AI agent personas defined via natural language
- Live WebRTC audio/video via LiveKit (SFU) with noise cancellation
- Real-time speech-to-text broadcast over data channels
- Post-meeting summaries powered by Gemini 2.5 Flash
- Calendar event creation with human-in-the-loop approval
- Role‑gated invite links with expiry management
- Subscription management via Polar.sh
Tech: Next.js 16, LiveKit, Neon DB + Drizzle ORM, Better Auth, Inngest, Gemini 2.5 Flash, tRPC, Tailwind CSS + shadcn/ui
Documentation · Repository
- Building AI‑powered web applications with real‑time capabilities
- Refining full‑stack architecture with Next.js App Router and serverless backends
- Contributing to open‑source and learning in public


