Production-ready AI backend platform featuring RAG, semantic search, RBAC, and AWS deployment
PolicyGuard AI is a production-ready Retrieval-Augmented Generation (RAG) platform for asking natural language questions about organizational documents. The platform combines semantic search with large language models to deliver accurate, context-aware answers while preserving source references, helping users get the information they need.
Key capabilities include:
• AI-powered document question answering (RAG)
• FastAPI asynchronous backend
• Semantic search with Qdrant Vector Database
• OpenAI GPT integration
• Redis response caching and chat memory
• Role-Based Access Control (RBAC)
• PDF & DOCX document ingestion pipeline
• AWS EC2 deployment with Docker and Nginx
• Production monitoring and health checks
Technologies: Python • FastAPI • OpenAI API • Qdrant • Redis • Docker • AWS EC2 • Nginx • Streamlit • Prometheus • Grafana
This project demonstrates end-to-end AI backend engineering, from document ingestion and semantic retrieval to deployment, monitoring, and production-ready APIs.