AI Agent Job Search - Multi-Platform Job Aggregator

A hackathon project that leverages AI agents and Model Context Protocol (MCP) to create an intelligent job search aggregator across multiple platforms including LinkedIn, Glassdoor, and other job sites.

🚀 Tech Stack

Backend

• Node.js with Express.js - RESTful API server
• Model Context Protocol (MCP) - AI agent framework
• Bright Data MCP Server - Web scraping and data extraction
• OpenAI GPT-4 - AI-powered job analysis and recommendations
• Winston - Logging and monitoring
• Axios - HTTP client for API calls
• Zod - Schema validation
• CORS - Cross-origin resource sharing

Frontend

• Next.js 15 - React framework with App Router
• React 19 - UI library
• TypeScript - Type safety
• Tailwind CSS - Utility-first CSS framework
• Heroicons - Icon library
• Headless UI - Accessible UI components

MCP Server

• Bright Data MCP Server (@brightdata/mcp) - Primary MCP server for web scraping
• FastMCP - MCP server framework
• Model Context Protocol SDK - MCP client implementation

🎯 Features

• Multi-Platform Job Search - Aggregates jobs from LinkedIn, Glassdoor, and other platforms
• AI-Powered Analysis - Uses GPT-4 to analyze job listings and provide insights
• Real-time Web Scraping - Leverages Bright Data's infrastructure for live data extraction
• Modern UI/UX - Clean, responsive interface built with Tailwind CSS
• Type Safety - Full TypeScript implementation for better development experience

🏗️ Architecture

The application follows a modern full-stack architecture:

┌─────────────────┐    ┌─────────────────┐    ┌─────────────────┐
│   Next.js       │    │   Express.js    │    │   Bright Data   │
│   Frontend      │◄──►│   Backend       │◄──►│   MCP Server    │
│   (React/TS)    │    │   (Node.js)     │    │   (Web Scraping)│
└─────────────────┘    └─────────────────┘    └─────────────────┘

🚀 Getting Started

Prerequisites

• Node.js 18+
• npm or yarn
• Bright Data API credentials
• OpenAI API key

Environment Variables

Create .env files in both backend/ and frontend/ directories:

# Backend .env
BRIGHTDATA_API_TOKEN=your_brightdata_token
BRIGHTDATA_GLASSDOOR_UNLOCKER_ZONE=your_glassdoor_zone
BRIGHTDATA_LINKEDIN_UNLOCKER_ZONE=your_linkedin_zone
BRIGHTDATA_CRUNCHBASE_UNLOCKER_ZONE=your_crunchbase_zone
BRIGHTDATA_NEWS_UNLOCKER_ZONE=your_news_zone
OPENAI_API_KEY=your_openai_key
CORS_ORIGIN=http://localhost:3000
PORT=3001

# Frontend .env
NEXT_PUBLIC_API_URL=http://localhost:3001

Installation

1. Backend Setup

cd backend
npm install
npm run dev

2. Frontend Setup

cd frontend
npm install
npm run dev

3. Start MCP Server

cd backend
node start-mcp.js

🎨 Hackathon Context

This project was built during a hackathon to demonstrate the power of AI agents and the Model Context Protocol (MCP) in creating intelligent applications. The goal was to showcase:

• AI Agent Integration - How MCP can connect AI agents to real-world data sources
• Web Scraping at Scale - Using Bright Data's infrastructure for reliable data extraction
• Modern Full-Stack Development - Combining cutting-edge frontend and backend technologies
• Real-time Data Processing - Aggregating and analyzing job data from multiple sources

🔧 API Endpoints

• GET /healthcheck - Health check endpoint
• GET /search?keyword=<term> - Search jobs by keyword
• POST /api/jobs/search - Advanced job search with filters

📁 Project Structure

ai-agent-job-search/
├── backend/
│   ├── src/
│   │   ├── mcp/           # MCP client and server integration
│   │   ├── routes/        # Express.js routes
│   │   ├── services/      # Business logic
│   │   └── server.js      # Main server file
│   └── package.json
├── frontend/
│   ├── app/
│   │   ├── components/    # React components
│   │   ├── api/          # Next.js API routes
│   │   └── page.tsx      # Main page
│   └── package.json
└── README.md

🤝 Contributing

This is a hackathon project, but contributions are welcome! Please feel free to submit issues or pull requests.

📄 License

This project is open source and available under the MIT License.

---

Built with ❤️ during a hackathon using cutting-edge AI and web technologies.