This project is a full-stack AI document Q&A application built using a Retrieval-Augmented Generation (RAG) architecture. Users can upload PDF documents and ask natural-language questions about their contents, with responses grounded in the uploaded material. The backend is built with Node.js and Express and uses LangChain to parse documents, generate embeddings, perform semantic retrieval, and construct context-aware prompts for a large language model.

The backend is containerized with Docker and deployed to AWS ECS (Fargate) behind an Application Load Balancer, with images stored in Amazon ECR. The frontend is built with React and TypeScript and styled using Chakra UI, providing a clean and intuitive interface for document ingestion and chat. The system follows a stateless, production-style architecture, demonstrating real-world experience with cloud deployment, container orchestration, and AI-powered application design.