Skip to main content

AI Agents Hackathon Project Recommender

AI Agents Next.js Python Docker MCP

Discover your perfect hackathon project with AI-powered GitHub profile analysis

An intelligent system that analyzes GitHub profiles and generates personalized hackathon project recommendations using AI agents, Model Context Protocol (MCP), and real-time data integration.

image

Features​

  • AI-Powered Analysis: Intelligent GitHub profile analysis using AI agents
  • Real-Time Data: Live integration with GitHub API through MCP servers
  • Personalized Recommendations: Tailored hackathon projects based on skills and interests
  • Secure Architecture: Containerized microservices with proper secrets management
  • Modern Stack: Next.js + Python + Docker + MCP integration
  • Production Ready: Scalable architecture with comprehensive tooling

Architecture​

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   Next.js UI    │───▶│  Python Agents  │───▶│   MCP Gateway   │
│   (Port 3003)   │    │   (Port 7777)   │    │   (Port 8811)   │
└─────────────────┘    └─────────────────┘    └─────────────────┘
         │                       │                       │
         │                       │                       ▼
         │                       │              ┌─────────────────┐
         │                       │              │  GitHub Tools   │
         │                       │              │  DuckDuckGo     │
         │                       │              │  Fetch Tools    │
         │                       │              │  (76 tools)     │
         │                       │              └─────────────────┘
         │                       ▼
         │              ┌─────────────────┐
         │              │   AI Models     │
         │              │   Qwen3 Small   │
         │              │   (Local/API)   │
         │              └─────────────────┘
         ▼
┌─────────────────┐
│     User        │
│   Browser       │
└─────────────────┘

Quick Start​

Prerequisites​

  • Docker & Docker Compose
  • GitHub Personal Access Token
  • 8GB+ RAM (for local AI models)

1. Clone & Setup​

git clone https://github.com/ajeetraina/ai-agents-hackathon-recommender
cd ai-agents-hackathon-recommender

2. Configure Secrets​

Create a file .mcp.env and add your GitHub PAT

nano .mcp.env
github.personal_access_token=XXX

3. Launch the Application​

# Build and start all services
docker compose up -d --build

# Monitor the logs
docker compose logs -f

4. Access the Application​

🎯 Usage​

  1. Open the web interface at http://localhost:3003
  2. Enter any GitHub username (e.g., "microsoft", "torvalds", "ajeetraina")
  3. Click "Get Recommendations"
  4. Receive personalized hackathon project suggestions!

🔧 Development​

Local Development Setup​

# Start in development mode
docker compose up --build

# Watch logs
docker compose logs -f agents-ui
docker compose logs -f agents

API Endpoints​

Agents Service (Port 7777)​

  • GET /health - Health check
  • POST /analyze - Analyze GitHub profile
  • GET /agents - List available agents
# Example API usage
curl -X POST http://localhost:7777/analyze \
  -H "Content-Type: application/json" \
  -d '{"username": "microsoft", "agent": "hackathon_recommender"}'