AIFitPro is an AI-powered fitness tracker application built with Java Spring Boot (Microservices), React, MongoDB, PostgreSQL, and Keycloak Authentication.
It leverages RabbitMQ for asynchronous communication and integrates with Gemini LLM to generate personalized fitness activity recommendations.

Below is the architecture diagram of AIFitPro:

• 🔑 Secure Authentication

  • Keycloak with PKCE flow for OAuth 2.0
  • Integrated Google OAuth2 login

• ⚡ Microservice Architecture

  • User Services (PostgreSQL backend)
  • Activity Services (MongoDB backend)
  • AI Services (MongoDB + Gemini LLM integration)

• 📡 Asynchronous Processing

  • RabbitMQ for handling async requests to the LLM model

• 🤖 AI-Powered Recommendations

  • Uses Gemini LLM to provide personalized fitness activity suggestions

• 🐳 Containerized Services

  • Dockerized Keycloak for authentication
  • Dockerized RabbitMQ for messaging

• Frontend: React.js
• Backend: Spring Boot (Microservices)
• Databases: MongoDB, PostgreSQL
• Authentication: Keycloak (PKCE + Google OAuth2)
• Message Queue: RabbitMQ
• AI Integration: Gemini LLM
• Service Discovery & Config: Eureka, Config Server
• Containerization: Docker

 git clone https://github.com/Jagdish2004/AIFitPro.git
cd AIFitPro