🔍 Project Overview

DiagnoSys is a modular, AI-powered system designed for interactive diagnostic support and smart information retrieval. It combines:

• FastAPI-based microservices for high-performance APIs,
• Large Language Models (LLMs) for conversational and reasoning capabilities,
• RAG (Retrieval-Augmented Generation) for enhanced, context-aware responses,
• Vector databases for fast and meaningful semantic search,
• Full Docker-based architecture for scalable deployment.

The architecture is divided into two key services:

• app/: Main diagnostic backend – includes chat, user handling, LLM interaction
• rag/: Retrieval engine – handles document indexing and semantic search

---

⚙️ Key Components

• app/: Main API service for diagnostic chat and user-facing endpoints
• rag/: Retrieval backend for document embedding and similarity search
• general_utils/: Text extraction utilities (e.g., from PDFs)
• test/: Basic pipeline testing
• Dockerfile.app / Dockerfile.rag: Container setup for API and retrieval services
• docker-compose.yml: Unified orchestration for both services