Joblume Backend

Joblume Backend is an AI-powered resume optimization and job matching system. The system uses advanced machine learning techniques to analyze resumes and job descriptions, providing intelligent matching and resume optimization capabilities.

🚀 Features

• Resume Analysis and Optimization using RAG (Retrieval Augmented Generation)
• Intelligent Job Description Matching
• Domain-Specific Knowledge Integration for Technical Roles
• Vector Database Integration (ChromaDB) for Efficient Information Retrieval
• RESTful API Backend (Django)

🧠 Technical Architecture

RAG-Based System

• Dynamic access to domain-specific knowledge
• Flexible response generation for various resume formats
• Enhanced relevance assessment for job-skill relationships
• Improved handling of technical terminology

Processing Pipeline

• PDF text extraction and processing
• Section identification and categorization
• Vector-based information retrieval
• Dynamic context incorporation

🛠️ Technology Stack

• Backend Framework: Django
• Machine Learning: RAG (Retrieval Augmented Generation)
• Vector Database: ChromaDB for efficient information storage and retrieval
• Package Manager: Poetry
• Text Processing:
  • sentence-transformers for text embeddings
  • PDF processing capabilities
• Core Focus: Computer Science and Technical Role Optimization

📁 Project Structure

.
├── ai/                     # AI components
│   ├── chromaDB.py        # ChromaDB integration
│   ├── embedder.py        # Text embedding functionality
│   ├── flatter.py        
│   └── rag.py             # RAG (Retrieval Augmented Generation)
│
├── FineTuning/            # Model fine-tuning
│   ├── train_model.py     # Model training script
│   ├── inference.py       # Inference implementation
│   ├── resume_processor.py # Resume processing utilities
│   └── results/           # Training checkpoints
│
├── job_backend/           # Django backend
│   ├── settings.py        # Project settings
│   ├── urls.py           # URL configurations
│   └── wsgi.py           # WSGI configurations
│
└── public/               # Public assets
    └── resumes/         # Resume storage

⚙️ Setup

Prerequisites

• Python >= 3.12
• Poetry >= 2.0.0

Dependencies

• Django >= 5.2.1
• sentence-transformers >= 4.1.0

1. Clone the repository

   git clone [repository-url]
   cd Joblume-back

2. Install dependencies using Poetry

   poetry install

3. Set up the environment

   poetry shell
   python manage.py migrate

4. Run the development server

   python manage.py runserver

🔧 Configuration

1. Environment Setup

   • Rename .env.example to .env (if provided)
   • Configure SECRET_KEY in Django settings
   • Set DEBUG = False in production
   • Configure ALLOWED_HOSTS for production deployment

2. Database Configuration

   • Default SQLite database for development
   • ChromaDB for vector storage and embeddings
   • Configure database settings in job_backend/settings.py

3. Django Settings

   • Time zone: UTC
   • Language: English (en-us)
   • Static files URL: /static/
   • Default auto field: BigAutoField

4. AI Model Configuration

   • Fine-tuned models are stored in FineTuning/resume_reorder_model/
   • Model checkpoints are available in FineTuning/results/
   • ChromaDB storage location: ai/chroma_store/

📚 API Documentation

The backend provides RESTful APIs for:

• Resume upload and analysis
• Job description matching
• Resume optimization suggestions

For detailed API documentation, refer to the Django endpoints defined in job_backend/urls.py.

🧪 Testing

The project includes test files and sample data:

• Test resumes in FineTuning/test/
• Training data in FineTuning/train/

📝 License

[Add License Information]

🤝 Contributing

[Add Contributing Guidelines]

📮 Contact

[Add Contact Information]