An advanced Medical AI Memory Agent system built with FastAPI (backend) and React + TanStack Query (frontend). The system maintains long-term conversational memory across sessions using a multi-layer memory architecture (conversational, semantic, entity, and summary memory).
It intelligently manages context windows by summarizing long conversations, expanding relevant memory when needed, and retrieving past context to improve response accuracy and continuity.
The frontend uses TanStack Query for efficient server-state management, caching, and synchronization with the backend memory system.
- React
- TanStack Query
- TailwindCSS (optional)
- FastAPI
- Pydantic
- PostgreSQL
- Pgvector
- Google Gemini
- Tavily API (for search tools)
- you can add and register new tool for the agent to use it
Make sure you have installed:
- Pgvector
- create vector extension
- documentation: https://github.com/pgvector/pgvector
git clone https://github.com/owolabi-develop/MemAgent.git
cd backend- GOOGLE_GEMINI_API_KEY
- DB_NAME
- TAVILY_API_KEY
- DB_PASSWORD
- DB_USER
- DB_HOST
- DB_PORT