A JavaScript-based chatbot implementation utilizing Retrieval-Augmented Generation (RAG) techniques for intelligent conversation handling.

This project is a retrieval-augmented generation chatbot that allows users to upload PDFs and ask questions using Google Gemini AI. The system processes uploaded documents by extracting text, converting it into vectors, storing these vectors in a database, and then retrieving relevant context to generate accurate and informative responses.

Key Features:

• PDF Upload: Users can upload PDF documents to serve as the knowledge base for the chatbot.
• Intelligent Q&A: Ask questions about the content of the uploaded PDFs and receive contextually relevant answers.
• RAG Implementation: Leverages Retrieval-Augmented Generation for enhanced response accuracy.
• AI Integration: Utilizes Google Gemini AI for natural language understanding and response generation.
• Vector Database: Stores document embeddings for efficient similarity search.

Tech Stack:

• Backend: FastAPI, Python 3.12, Sentence Transformers, Google Gemini, Supabase, Uvicorn.
• Frontend: React 19, Vite, Axios, Tailwind CSS.

Setup: The project includes detailed setup instructions for cloning, installing dependencies, configuring environment variables (including Supabase credentials and Gemini API key), and running both the backend and frontend applications.

Usage: Simply upload a PDF, type your question, and the chatbot will provide answers based on the document's content.