Synapse - Context-Aware Enterprise AI

Synapse is a production-ready, context-aware Retrieval-Augmented Generation (RAG) system. It is designed for high-performance enterprise document search with sub-200ms latency.

Architecture

The system is built as a microservices-oriented architecture:

• Frontend: React + TypeScript SPA with real-time SSE streaming.
• API Gateway: FastAPI orchestrator managing Gemini API and MCP tool calls.
• Retrieval Engine: PyTorch-based document embedder with FAISS vector search.

Tech Stack

• AI: Gemini API, PyTorch, FAISS, Model Context Protocol (MCP).
• Backend: FastAPI, Uvicorn, Python 3.10.
• Frontend: React, TypeScript, Vite, Tailwind-style CSS.
• Orchestration: Docker, Docker Compose.

Getting Started

1. Set your Gemini API key in a .env file:

   GEMINI_API_KEY=your_key_here

2. Start the services:

   docker-compose up --build