An end-to-end Retrieval-Augmented Generation (RAG) based customer support automation platform that intelligently answers customer queries using a knowledge base and automates Gmail responses.
This project automates customer support by combining semantic search with context-aware response generation.
User queries are matched against a domain knowledge base using vector embeddings, and accurate responses are generated and sent via Gmail automation.
- 🔍 Semantic Knowledge Retrieval (RAG)
- 🧠 Context-Aware Response Generation
- 📚 ChromaDB Persistent Vector Store
- 📩 Automated Gmail Draft Creation & Sending
- ⚡ FastAPI Backend with Swagger UI
- 🧩 Multiple Indexing Options (Chroma / sklearn / FAISS)
User Query ↓ FastAPI Endpoint ↓ Vector Retrieval (ChromaDB / sklearn) ↓ Context Injection (RAG) ↓ Response Generation ↓ Gmail Draft / Email Send