RecruitAI - Intelligent Recruitment System

RecruitAI is a next-generation Applicant Tracking System (ATS) designed to streamline the hiring process. By combining the MERN Stack with a dedicated Python Machine Learning microservice and Google Gemini, it automatically parses resumes, scores candidates based on job descriptions, and provides visual analytics to help HR make data-driven decisions.

Key Features

• Smart Resume Parsing: Automatically extracts text and skills from PDF resumes using OCR and NLP.
• Hybrid AI Scoring:
  • ML Model: Uses a custom Python-trained model (recruit_model.pkl) to score candidates based on historical dataset averages.
  • GenAI Integration: Utilizes Google Gemini API for semantic understanding and qualitative "culture fit" analysis.
• Interactive Dashboard: Visual analytics using Recharts to display candidate leaderboards and skill gap analysis.
• Job Role Configuration: Dynamic job setup allows recruiters to define specific weighting for skills and experience.
• Secure Authentication: JWT-based authentication with encrypted user data.

Application Screenshots

1. Secure Authentication

Secure login portal for HR administrators.

2. Dashboard Overview

A comprehensive view of active job postings and recent applicants.

3. Job Configuration

Dynamic setup to define required skills and weightage for specific roles.

4. Candidate Analysis & Leaderboard

Detailed breakdown of candidate scores, resume parsing results, and AI-generated insights.

5. Model Tuning & Analytics

Visualization of the underlying ML scoring metrics and dataset averages.

---

Tech Stack

Frontend (Client)

• Framework: React (Vite)
• Styling: Tailwind CSS
• State Management: React Context API
• Visualizations: Recharts
• HTTP Client: Axios

Backend (Server)

• Runtime: Node.js & Express.js
• Database: MongoDB (Mongoose)
• Authentication: JWT (JSON Web Tokens)
• Resume Parsing: PDF-parse / Multer

Machine Learning Service

• Language: Python 3.13
• Framework: Flask (Microservice)
• Models: Scikit-learn (Pickle), Google Gemini API
• Data Processing: NumPy, Pandas

Project Structure

.
├── client/          # React Frontend application
├── server/          # Node.js/Express Backend API
├── ml_service/      # Python Machine Learning Microservice
├── demo_resume/     # Sample resumes for testing
└── screenshots/     # Application preview images

Getting Started

Follow these instructions to set up the project locally.

Prerequisites

• Node.js (v18+)
• Python (v3.10+)
• MongoDB (Local or Atlas URL)

1. Backend Setup (Node.js)

cd server
npm install
# Create a .env file based on the example below
npm start

2. Frontend Setup (React)

cd client
npm install
npm run dev

3. ML Service Setup (Python)

cd ml_service
python -m venv venv

# Windows
.\venv\Scripts\activate

# Mac/Linux
source venv/bin/activate

pip install -r requirements.txt
python app.py

Environment Variables

Create a .env file in the server directory:

PORT=5000
MONGO_URI=your_mongodb_connection_string
JWT_SECRET=your_jwt_secret
GEMINI_API_KEY=your_google_gemini_key
ML_SERVICE_URL=http://localhost:8000

How It Works

1. Job Setup: The recruiter creates a job profile (e.g., "Senior React Dev") and defines required skills.

2. Upload: Resumes (PDF) are uploaded via the frontend.

3. Processing:

   • The Node Backend saves the file and extracts raw text.
   • Text is sent to the Python Microservice.
   • The ML Model vectorizes the resume text and calculates a similarity score against the job description.
   • Google Gemini provides a secondary qualitative summary.

4. Results: The frontend displays a ranked leaderboard of candidates with detailed score breakdowns.

Contributing

Contributions are welcome! Please fork the repository and submit a pull request.

License

This project is licensed under the MIT License.