Enterprise Document Chatbot

This project demonstrates the implementation of a Retrieval-Augmented Generation (RAG) chatbot capable of processing and answering queries from various enterprise document formats including PDF and DOCX files. The solution leverages Ollama for local LLM inference, eliminating dependencies on external APIs while maintaining high performance and data privacy.

• Developed a RAG chatbot capable of answering queries from enterprise documents (PDF, DOCX).
• Integrated Ollama for local LLM inference, eliminating reliance on external APIs (OpenAI/Claude).
• Containerized services with Docker Compose and exposed REST endpoints for ingestion, health, and query.
• Delivered a full-stack solution (React frontend + FastAPI backend) demonstrating scalable and modular AI deployment.

• Implementing efficient document parsing and chunking strategies for optimal retrieval
• Managing vector embeddings and similarity search with ChromaDB
• Optimizing local LLM performance while maintaining response quality
• Creating a robust Docker-based architecture for easy deployment