A Retrieval-Augmented Generation (RAG) Chatbot powered by LangChain, GPT-4o, and a modern web stack. It features a FastAPI backend, a MongoDB-integrated chat history, and a Next.js frontend styled with TailwindCSS and DaisyUI.
This chatbot allows users to ask questions based on documents ingested into a vector store and answered using GPT-4o via LangChain. The FastAPI backend manages RAG queries and logs chat history into MongoDB, while the Next.js frontend provides an interactive, responsive chat UI.
- GPT-4o powered answers using LangChain agents
- PDF/text ingestion and chunking for context-aware retrieval
- Vector-based similarity search
- Chat history persistence using MongoDB
- Session-based unique chat identification
- FastAPI backend with async handling
- Next.js frontend with TailwindCSS + DaisyUI for modern UI/UX