Resume Parser API 🚀

A powerful AI-driven FastAPI application that extracts structured information from PDF and Word resume documents. This application can be deployed on Vercel as a serverless function.

✨ Features

• Multi-format Support: Parse PDF, DOC, and DOCX resume files
• AI-Powered Extraction: Extract key information using advanced NLP techniques
• Structured Data Output: Returns clean, structured JSON data
• Batch Processing: Process multiple resumes simultaneously
• RESTful API: Clean, documented REST API endpoints
• Serverless Ready: Optimized for Vercel serverless deployment
• CORS Enabled: Ready for frontend integration

📋 Extracted Information

The API extracts the following information from resumes:

• Personal Information: Name, contact details
• Contact Information: Email addresses, phone numbers, LinkedIn profiles, location
• Skills: Technical skills with categorization and relevance scoring
• Education: Degrees, institutions, graduation years
• Work Experience: Job titles, companies, years of experience
• Professional Summary: Extracted objective/summary sections

🏗️ Project Structure

resume-parser-api/
├── main.py              # FastAPI application entry point
├── resume_parser.py     # Core resume parsing logic
├── requirements.txt     # Python dependencies
├── vercel.json         # Vercel deployment configuration
└── README.md           # This file

🚀 Quick Start

Local Development

1. Clone the repository

   git clone <your-repo-url>
   cd resume-parser-api

2. Create virtual environment

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Download spaCy model (optional but recommended)

   python -m spacy download en_core_web_sm

5. Run the application

   uvicorn main:app --reload

The API will be available at http://localhost:8000

🌐 Deployment on Vercel

Prerequisites

• GitHub account
• Vercel account (free tier available)
• Git installed locally

Step-by-Step Deployment

1. Push code to GitHub

   git init
   git add .
   git commit -m "Initial commit: Resume Parser API"
   git branch -M main
   git remote add origin <your-github-repo-url>
   git push -u origin main

2. Deploy to Vercel

   Option A: Using Vercel Dashboard

   • Go to vercel.com
   • Click "New Project"
   • Import your GitHub repository
   • Vercel will automatically detect the Python project
   • Click "Deploy"

   Option B: Using Vercel CLI

   npm install -g vercel
   vercel login
   vercel

3. Environment Variables (if needed)

   • In Vercel dashboard, go to Project Settings > Environment Variables
   • Add any required environment variables

Important Notes for Vercel Deployment

• File Size Limit: Vercel has a 250MB unzipped size limit for serverless functions
• Execution Time: Maximum 30 seconds per request on free tier
• Memory Limit: 1024MB on free tier
• Cold Starts: First requests might be slower due to cold starts

📡 API Endpoints

1. Health Check

GET /
GET /health

Response:

{
  "message": "Resume Parser API is running",
  "version": "1.0.0",
  "status": "healthy",
  "supported_formats": ["PDF", "DOC", "DOCX"]
}

2. Parse Single Resume

POST /parse-resume
Content-Type: multipart/form-data

Request: Upload a resume file (PDF, DOC, or DOCX)

Response:

{
  "status": "success",
  "message": "Resume parsed successfully",
  "filename": "john_doe_resume.pdf",
  "file_size": 245760,
  "data": {
    "personal_info": {
      "name": "John Doe"
    },
    "contact_info": {
      "emails": ["john.doe@email.com"],
      "phones": ["+1-555-123-4567"],
      "linkedin": "john-doe-dev",
      "location": ["San Francisco, CA"]
    },
    "skills": {
      "technical_skills": [
        {
          "skill": "Python",
          "mentions": 5,
          "category": "programming_language"
        }
      ]
    },
    "education": {
      "degrees": [
        {
          "degree": "Bachelor",
          "field": "Computer Science",
          "type": "degree"
        }
      ]
    },
    "experience": {
      "total_years_experience": 3,
      "experience_level": "junior"
    }
  }
}

3. Parse Multiple Resumes

POST /parse-resume-batch
Content-Type: multipart/form-data

Request: Upload up to 10 resume files

Response:

{
  "status": "success",
  "message": "Processed 3 files",
  "results": [
    {
      "filename": "resume1.pdf",
      "status": "success",
      "data": { ... }
    }
  ]
}

🧪 Testing the API

Using curl

# Health check
curl https://your-vercel-url.vercel.app/

# Parse resume
curl -X POST https://your-vercel-url.vercel.app/parse-resume \
  -F "file=@path/to/your/resume.pdf"

Using Python

import requests

# Parse resume
with open('resume.pdf', 'rb') as f:
    response = requests.post(
        'https://your-vercel-url.vercel.app/parse-resume',
        files={'file': f}
    )
    print(response.json())

Using JavaScript/Frontend

const formData = new FormData();
formData.append('file', fileInput.files[0]);

fetch('https://your-vercel-url.vercel.app/parse-resume', {
    method: 'POST',
    body: formData
})
.then(response => response.json())
.then(data => console.log(data));

🔧 Configuration

File Size Limits

• Maximum file size: 10MB per file
• Maximum files in batch: 10 files

Supported File Types

• PDF (application/pdf)
• Word Document (application/vnd.openxmlformats-officedocument.wordprocessingml.document)
• Legacy Word Document (application/msword)

📈 Performance Optimization

• Lazy Loading: NLP models are loaded only when needed
• Text Caching: Extracted text is cached during processing
• Async Processing: All operations are asynchronous
• Memory Management: Large files are processed in streams

🐛 Troubleshooting

Common Issues

1. "Module not found" errors

   • Ensure all dependencies in requirements.txt are correctly specified
   • Check Python version compatibility

2. File upload fails

   • Verify file size is under 10MB
   • Check file format is supported
   • Ensure proper Content-Type headers

3. Vercel deployment fails

   • Check that vercel.json configuration is correct
   • Verify file structure matches expected layout
   • Review Vercel build logs for specific errors

4. spaCy model not found

   • The app will work without spaCy but with reduced accuracy
   • For production, consider pre-downloading the model

Error Codes

• 400: Bad Request (unsupported file type, file too large)
• 413: Payload Too Large (file exceeds 10MB)
• 500: Internal Server Error (processing error)

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

📄 License

This project is licensed under the MIT License.

🙏 Acknowledgments

• Built with FastAPI
• PDF processing powered by pdfplumber
• Word document processing with python-docx
• NLP capabilities via spaCy
• Deployed on Vercel

---

Made with ❤️ for the developer community