Build a RAG (Retrieval-Augmented Generation) over your own PDFs from a web UI: upload documents, choose chunking and embedding settings, and get a dedicated API endpoint + API key to query your knowledge base from any app. Add files or re-index ("update memory") at any time from the dashboard.

• frontend/ — Next.js (App Router, TypeScript, Tailwind, shadcn/ui). Auth via Supabase.
• backend/ — FastAPI. Ingestion (PyMuPDF → chunking → embeddings → pgvector), retrieval, generation, API-key management.
• supabase/ — SQL migrations: Postgres metadata tables, pgvector chunks, RLS policies, private storage bucket.
• legacy/ — the original prototype, kept for reference.

Providers: OpenAI (embeddings + chat), Ollama (local, optional), sentence-transformers (local embeddings, optional).

Create a project at supabase.com, then run the SQL in supabase/migrations/0001_init.sql and 0002_rls.sql (SQL Editor, in order).

cd backend
py -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt
copy .env.example .env   # then fill in the values
uvicorn app.main:app --reload

API runs at http://localhost:8000 (interactive docs at /docs).

Optional local embeddings (downloads PyTorch, ~2.5 GB):

pip install -r requirements-local.txt

Optional local LLM: install Ollama and pull a model, e.g. ollama pull llama3.1 and ollama pull nomic-embed-text.

cd frontend
npm install
copy .env.example .env.local   # then fill in the values
npm run dev

App runs at http://localhost:3000.

Every project gets API keys (Project → API tab). Query it from anywhere:

curl -X POST http://localhost:8000/v1/projects/<project-id>/query \
  -H "Authorization: Bearer oreag_sk_..." \
  -H "Content-Type: application/json" \
  -d '{"question": "What does the handbook say about onboarding?"}'

Response: { "answer": "...", "sources": [{ "filename", "page_number", "content", "similarity" }], "model", "latency_ms" }