Resume Parser System

A comprehensive resume parsing system that extracts structured data from PDF resumes using LangChain and OpenAI, and stores the data in a SQLite database categorized by user.

Features

• 📄 PDF Text Extraction: Extract text from PDF resume files using PyPDF2
• 🤖 AI-Powered Parsing: Use LangChain with OpenAI to intelligently parse resume content
• 🗃️ Structured Data Storage: Store parsed data in SQLite database with user categorization
• 👤 User Management: Automatically categorize resumes by user name (inferred or specified)
• 📊 Comprehensive Data Extraction: Extract education, experience, skills, projects, achievements, and more

Installation

1. Clone or set up the project
2. Install required dependencies:

pip install -r requirements.txt

3. Set up your OpenAI API credentials in a .env file:

OPENROUTER_API_KEY=your_api_key_here
OPENROUTER_BASE_URL=https://openrouter.ai/api/v1
YOUR_SITE_URL=your_site_url
YOUR_SITE_NAME=your_site_name

Usage

1. Add Resume Files

Place your PDF resume files in the resume/ folder:

resume/
├── arpit_solanki_resume.pdf
├── john_doe_cv.pdf
└── jane_smith_resume.pdf

2. Process Resumes

Process All Resumes

python main.py

This will process all PDF files in the resume/ folder and automatically infer user names from the file content or filename.

Process All Resumes for a Specific User

python main.py --user "Arpit Solanki"

Process a Specific File

python main.py --file "resume.pdf" --user "John Doe"

List All Users and Resumes

python main.py --list

3. Command Line Options

• --user, -u: Specify user name to associate with resume(s)
• --file, -f: Process a specific PDF file
• --list, -l: List all users and their resumes in the database

Project Structure

langchain/
├── resume/                 # Folder for PDF resume files
├── main.py                # Main script to run the resume parser
├── pdf_parser.py          # PDF text extraction module
├── resume_parser.py       # LangChain-based resume parsing module
├── database.py            # Database models and management
├── model.py               # Original LangChain model configuration
├── requirements.txt       # Python dependencies
├── .env                   # Environment variables (API keys)
└── README.md             # This file

Database Schema

Users Table

• id: Primary key
• name: User name
• created_at: Timestamp

Resumes Table

• id: Primary key
• user_id: Foreign key to users table
• file_name: Original PDF filename
• full_name: Extracted full name
• email: Email address
• phone: Phone number
• location: Current location
• linkedin: LinkedIn profile URL
• github: GitHub profile URL
• summary: Professional summary
• education: JSON string of education details
• experience: JSON string of work experience
• technical_skills: JSON string of technical skills
• soft_skills: JSON string of soft skills
• projects: JSON string of projects
• achievements: JSON string of achievements
• certifications: JSON string of certifications
• raw_text: Original extracted text
• created_at: Timestamp
• updated_at: Last update timestamp

Extracted Data Structure

The system extracts the following structured data from resumes:

Personal Information

• Full name
• Email address
• Phone number
• Current location
• LinkedIn profile
• GitHub profile

Professional Summary

• Brief professional summary or objective

Education

• Institution name
• Degree/qualification
• Field of study
• Graduation year
• GPA (if mentioned)
• Location

Work Experience

• Company name
• Position/job title
• Duration of employment
• Location
• Key responsibilities and achievements

Skills

• Technical skills (programming languages, frameworks, tools)
• Soft skills

Projects

• Project name
• Description
• Technologies used
• Duration
• Project links

Achievements & Awards

• Achievement title
• Description
• Year achieved
• Awarding organization

Certifications

• Certification name
• Issuing organization
• Issue date
• Expiry date
• Credential ID

Example Usage

from pdf_parser import PDFParser
from resume_parser import ResumeParser
from database import DatabaseManager

# Initialize components
pdf_parser = PDFParser()
resume_parser = ResumeParser()
db_manager = DatabaseManager()

# Extract text from a PDF
text = pdf_parser.extract_text_from_pdf("resume/arpit_resume.pdf")

# Parse the resume
parsed_data = resume_parser.parse_resume(text)

# Save to database
resume_record = db_manager.save_resume_data("Arpit Solanki", "arpit_resume.pdf", parsed_data)

Error Handling

The system includes comprehensive error handling:

• Failed PDF text extraction
• AI parsing failures (falls back to regex-based extraction)
• Database connection issues
• File not found errors

Output Example

🚀 Starting Resume Processing Pipeline
=====================================
🔧 Initializing components...
📄 Found 1 PDF file(s): ['arpit_resume.pdf']

============================================================
Processing: arpit_resume.pdf
============================================================
✅ Successfully extracted text (2847 characters)
🔄 Parsing resume with AI...
👤 User identified as: Arpit Solanki
💾 Saving to database...
✅ Successfully saved resume data for Arpit Solanki
📝 Resume ID: 1

📊 Extracted Data Summary:
   • Full Name: Arpit Solanki
   • Email: arpitsolanki6825@gmail.com
   • Phone: +91-8279824227
   • Location: N/A
   • Education entries: 1
   • Work experience entries: 1
   • Projects: 2
   • Technical skills: 8

🎯 Processing Complete!
========================================
✅ Successfully processed: 1
❌ Failed to process: 0
📊 Total files: 1

💾 Data saved to: resume_database.db
🔍 You can query the database to retrieve user resume data

Contributing

Feel free to contribute by:

• Adding support for more file formats (DOC, DOCX)
• Improving the AI parsing prompts
• Adding more structured data fields
• Enhancing error handling
• Adding a web interface

License

This project is open source. Feel free to use and modify as needed.