RAG · Vector search · LLM · Open source

Ask anything
about coffee.

A RAG (Retrieval-Augmented Generation) system built on top of The World Atlas of Coffee by James Hoffmann. Ask anything — from origin regions and processing methods to flavor profiles and brewing techniques.

How it works

01 your question is embedded into a vector
02 FAISS finds the most relevant book excerpts
03 Llama 3.3 70B generates a grounded answer

Built with

Python FastAPI FAISS all-MiniLM-L6-v2 Llama 3.3 70B Groq API

Hey! I'm a coffee expert trained on The World Atlas of Coffee. Ask me anything — origins, processing, brewing, flavor profiles... ☕

Please ask your questions in English only.

⚠ free instance spins down with inactivity — first answer can take 50s+