Transform conversations into actionable insights.
An AI-powered meeting intelligence platform that goes beyond transcription: Briefly automatically summarizes meetings, extracts action items and decisions, tracks deadlines, and lets you chat with your meeting history using Retrieval-Augmented Generation (RAG).
Built as a production-grade, portfolio-quality full-stack project — React + TypeScript on the frontend, FastAPI on the backend, Neon Postgres with pgvector for semantic search, Supabase for auth and storage, and Groq for fast Whisper transcription and LLM inference. Designed to run entirely on free-tier infrastructure.
Actively in development, built with full documentation at each stage.
- Frontend: React, TypeScript, Vite, TailwindCSS — deployed to Vercel
- Backend: FastAPI, SQLAlchemy (async), Alembic — deployed to Render (free tier)
- Database: Neon PostgreSQL with pgvector
- Auth & Storage: Supabase
- AI: Groq (Whisper STT + LLM), sentence-transformers (embeddings)
cd backend
python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .env
uvicorn app.main:app --reloadcd frontend
npm install
cp .env.example .env
npm run devVisit http://localhost:5173 — it should show "Backend status: connected".
Built incrementally, one milestone at a time. See docs/ for the full architectural review, SRS, and milestone roadmap (Milestones 1–12 core, Milestone 13 optional speaker diarization).