CV Analysis System

A comprehensive system for processing, analyzing, and querying CV/resume documents using OCR and LLMs. Transform your recruitment workflow with AI-powered candidate analysis.

• Document Processing: Handle PDF and Word documents with advanced OCR capabilities
• Information Extraction: Extract structured data from CVs using AI models
• Natural Language Querying: Ask questions about candidates in plain English
• Web Interface: User-friendly Streamlit application with multiple functionality tabs
• Multi-model Support: Works with Claude, GPT, and Gemini models
• Interactive CV Viewing: View extracted CV information in structured format
• Conversation Memory: System remembers context for follow-up questions
• Profile Management: Create and manage your own professional profile
• Job Analysis: Analyze job descriptions and generate matching scores
• Cover Letter Generation: Auto-generate customized cover letters
• Resume Editor: Interactive interface for updating resume content

• Time Savings: Reduce CV review time by up to 75% through automated information extraction
• Consistent Evaluation: Standardize candidate information for fair comparison
• Better Matching: Quickly identify candidates with specific skills or experience
• Insights Discovery: Uncover patterns and connections in candidate data through natural language queries

• Centralized Database: Keep all candidate information in one organized system
• Collaborative Hiring: Share insights with team members through the intuitive interface
• Reduced Bias: Focus on skills and qualifications with structured data
• Efficient Screening: Pre-screen candidates at scale with intelligent queries

• Improved Hiring Efficiency: Decrease time-to-hire by streamlining CV analysis
• Better Talent Identification: Discover ideal candidates that might be overlooked in manual processes
• Scalable Recruitment: Process hundreds of CVs quickly during high-volume hiring periods
• Data-Driven Decisions: Base hiring decisions on comprehensive candidate analysis

• Profile Management: Maintain your professional profile and resume in one place
• Job Match Analysis: Get detailed analysis of your fit for specific job postings
• Automated Cover Letters: Generate customized cover letters based on job descriptions
• Resume Enhancement: Edit and improve your resume with an interactive interface
• Application History: Track your job applications and analyses

• Python 3.8+
• Tesseract OCR
• LLM API key (Anthropic Claude, OpenAI GPT, or Google Gemini)
• Poppler (for PDF processing)

1. Clone the repository:

git clone https://github.com/MohiuddinSumon/cv_analyzer.git
cd cv_analyzer

2. Create and activate a virtual environment:

   Using venv (Python's built-in module)

   # Create virtual environment
   python -m venv .venv
   
   # Activate virtual environment
   # On Windows
   .venv\Scripts\activate
   
   # On macOS/Linux
   source .venv/bin/activate

   Using virtualenv (alternative)

   # Install virtualenv if not already installed
   pip install virtualenv
   
   # Create virtual environment
   virtualenv .venv
   
   # Activate virtual environment
   # On Windows
   .venv\Scripts\activate
   
   # On macOS/Linux
   source .venv/bin/activate

   Your command prompt should now show the virtual environment name, indicating it's active.

3. Install Python dependencies:

pip install -r requirements.txt

4. Install Tesseract OCR:

   • Ubuntu: sudo apt-get install tesseract-ocr poppler-utils
   • macOS: brew install tesseract poppler
   • Windows:
     • Download Tesseract from GitHub
     • Add Tesseract to your PATH environment variable
     • Download Poppler binaries for Windows

5. Create a .env file with your API keys (choose one):

ANTHROPIC_API_KEY=your_anthropic_api_key
# or
OPENAI_API_KEY=your_openai_api_key
# or
GOOGLE_API_KEY=your_google_api_key

1. Ensure your virtual environment is activated (you should see .venv in your command prompt)

2. Run the application:

streamlit run app.py

3. Open your browser and navigate to the provided URL (typically http://localhost:8501)

4. Upload CV files through the sidebar

5. Process the CVs using the "Process CVs" button

6. Use the various tabs for different functionalities:

• Query CV information using natural language
• View conversation history
• Get insights about candidates

• View structured CV information
• Access detailed candidate profiles
• Review extracted data in JSON format

• Upload and manage your personal resume
• Add professional links (Portfolio, GitHub, LinkedIn)
• Include additional information
• View analyzed profile information

• Input job descriptions manually or via URL
• Get detailed job match analysis
• Generate customized cover letters
• Track application history
• Download analyses and cover letters
• Compare your profile against job requirements

• Edit personal information interactively
• Update skills and experience
• Modify work history details
• Save changes in real-time

7. View detailed CV information by clicking on a candidate name in the sidebar

8. When you're done, deactivate the virtual environment:

deactivate

cv-analysis-system/
├── app.py                  # Main application file
├── requirements.txt        # Python dependencies
├── .env                    # Environment variables (API keys)
├── .venv/                  # Virtual environment directory
├── cv_database.json        # Database file for stored CV data
├── data
│    └── sample_cvs/             # Sample CV files for testing
│        ├── candidate1.pdf
│        └── candidate2.docx
└── README.md               # This file

1. Document Processing:

   • Extracts text from PDFs and Word documents
   • Uses OCR for scanned documents or images with adaptive thresholding
   • Handles complex document layouts and formatting

2. Information Extraction:

   • Sends CV text to LLM (Claude, GPT, or Gemini)
   • Uses prompt engineering to extract structured information
   • Organizes data into categories:
     • Personal information (name, contact details)
     • Education history
     • Work experience
     • Skills (technical, soft, languages)
     • Projects
     • Certifications

3. Data Storage:

   • Stores structured CV data in a JSON database
   • Maintains efficient indexing for quick retrieval
   • Persists data between application sessions

4. Query Interface:

   • Uses LLM to interpret natural language queries about candidates
   • Maintains conversation context for follow-up questions
   • Provides informative responses based on candidate data
   • Supports comparison of multiple candidates

5. CV Viewer:

   • Displays structured CV information in an easy-to-read format
   • Allows quick navigation between different candidates
   • Presents complete candidate profiles

The system uses a modular architecture with several key components:

• CVProcessor: Handles document processing with fallback methods for difficult documents
• CVAnalyzer: Extracts structured information using LLM models
• CVDatabase: Manages data storage and retrieval
• CVQueryEngine: Processes natural language queries about candidates
• Streamlit Interface: Provides a user-friendly web interface

• OCR Engine: You can modify the OCR engine in the CVProcessor class
• LLM Provider: Choose between Anthropic Claude, OpenAI GPT, or Google Gemini
• Database Storage: Change the storage format in the CVDatabase class
• UI Customization: Extend the Streamlit interface to add custom features

• OCR processing is resource-intensive for large documents
• LLM API calls incur costs based on your provider's pricing
• Processing speed depends on document complexity and selected models

• CV similarity scoring
• Candidate recommendation engine
• Skills gap analysis
• Integration with applicant tracking systems
• Export to common HR formats

MIT

Contributions are welcome! Please feel free to submit a Pull Request.