An AI-powered serverless application that automatically extracts information from resumes and matches candidates with relevant job opportunities using AWS services.

This system automates the resume screening process by:

• Extracting text from PDF resumes using AWS Textract (99% accuracy)
• Parsing skills, experience, and contact information
• Matching candidates with jobs based on skill overlap
• Providing a user-friendly dashboard for recruiters

Processing Time: 8-10 seconds end-to-end | Cost: ~$0.01 per resume

• 📄 Automated Text Extraction - AWS Textract extracts text with 99% confidence
• 🧠 Intelligent Skill Parsing - Identifies skills, experience, and contact info
• 🎯 Smart Job Matching - Calculates match scores with skill gap analysis
• 📊 Recruiter Dashboard - Clean UI for viewing candidates and matches
• ⚡ Real-time Processing - Event-driven architecture with auto-triggers
• 🔄 Fully Automated - No manual intervention required
• 📈 Scalable - Serverless architecture handles unlimited concurrent uploads

User Upload (PDF)
    ↓
API Gateway (/upload)
    ↓
Upload Handler Lambda → S3 (resumes/raw/)
    ↓ [S3 Event Trigger]
Textract Processor Lambda → S3 (resumes/processed/)
    ↓ [S3 Event Trigger]
Skill Matcher Lambda → DynamoDB (CandidateProfiles)
    ↓ [DynamoDB Stream Trigger]
Job Matcher Lambda → DynamoDB (Matches)
    ↓
Dashboard (API Gateway /candidates & /matches)

Frontend:

• HTML5, CSS3, JavaScript (Vanilla)
• Responsive design

Backend:

• AWS Lambda (Python 3.9)
• AWS API Gateway (REST API)
• AWS S3 (Object Storage)
• AWS DynamoDB (NoSQL Database)
• AWS Textract (AI Text Extraction)

DevOps:

• AWS IAM (Access Management)
• AWS CloudWatch (Logging & Monitoring)
• Git/GitHub (Version Control)

• AWS Account with appropriate permissions
• AWS CLI configured
• Python 3.9+
• Git

git clone https://github.com/Anshuman-git-code/Resume-Parser-Skill-Matcher.git
cd Resume-Parser-Skill-Matcher

Create S3 Bucket:

bash scripts/setup-s3.sh

Deploy Lambda Functions:

bash scripts/deploy-lambdas.sh

Setup API Gateway:

bash scripts/setup-api-gateway.sh

open frontend/dashboard/index.html

Or start a local server:

cd frontend/dashboard
python3 -m http.server 8000
# Open http://localhost:8000

1. Open the dashboard
2. Click "Choose File" and select a PDF resume
3. Wait 8-10 seconds for processing
4. View extracted text in the popup modal

1. Click the "Candidates" tab
2. Browse all processed candidates
3. Search by name or skills
4. Click "View Matches" for any candidate

1. Click the "Matches" tab
2. Select a candidate from the dropdown
3. See ranked job matches with scores
4. Review matched and missing skills

Resume-Parser-Skill-Matcher/
├── backend/
│   ├── lambda-functions/
│   │   ├── upload-handler/          # Handles resume uploads
│   │   ├── textract-processor/      # Extracts text from PDFs
│   │   ├── skill-parser/            # Parses skills and experience
│   │   ├── job-matcher/             # Matches candidates with jobs
│   │   ├── get-candidates/          # API to fetch candidates
│   │   └── get-matches/             # API to fetch job matches
│   ├── tests/                       # Test files
│   └── utils/                       # Utility functions
├── frontend/
│   ├── dashboard/
│   │   ├── index.html               # Main dashboard
│   │   ├── app.js                   # Business logic
│   │   ├── styles.css               # Styling
│   │   └── check-status.html        # Status checker
│   └── assets/                      # Static assets
├── scripts/
│   ├── setup-s3.sh                  # S3 bucket setup
│   ├── deploy-lambdas.sh            # Lambda deployment
│   ├── setup-api-gateway.sh         # API Gateway setup
│   ├── trigger-job-matching.sh      # Manual match trigger
│   ├── validate-setup.sh            # Validate AWS setup
│   ├── test-upload.py               # Test upload script
│   └── create-dateutil-layer.sh     # Create Lambda layer
├── docs/
│   └── handoff-notes/               # Documentation
├── PROJECT_REPORT.md                # Comprehensive project report
├── requirements.txt                 # Python dependencies
└── README.md                        # This file

Base URL: https://n8ujlpobn6.execute-api.ap-south-1.amazonaws.com/prod

• POST /upload - Upload resume
• GET /candidates - Fetch all candidates
• GET /matches?candidate_id=xxx - Fetch job matches

Update these in frontend/dashboard/app.js:

const API_ENDPOINT = 'YOUR_API_GATEWAY_URL/prod/upload';
const CANDIDATES_API = 'YOUR_API_GATEWAY_URL/prod/candidates';
const MATCHES_API = 'YOUR_API_GATEWAY_URL/prod/matches';

bash scripts/validate-setup.sh

python3 scripts/test-upload.py path/to/resume.pdf

1. Upload a test resume through the dashboard
2. Check S3 bucket for uploaded files
3. Verify DynamoDB tables have data
4. Check CloudWatch logs for any errors

The system comes with sample data:

• 9 candidates processed
• 10+ jobs in database
• 100+ matches generated

Sample candidate:

{
  "resume_id": "818cee79-8493-4f9a-ae57-b9af695e82b2",
  "name": "TANNU NAGAR",
  "email": "[email protected]",
  "skills": ["docker", "kubernetes", "python", "aws"],
  "total_experience_years": 5.2
}

• Only supports PDF files (no Word documents)
• Basic pattern matching for skill extraction
• Simple skill overlap matching algorithm
• No user authentication
• No duplicate resume detection
• Limited to predefined skill list

See PROJECT_REPORT.md for detailed limitations.

• AWS Comprehend integration for better NLP
• Support for Word documents and images
• User authentication with AWS Cognito
• Enhanced matching algorithm

• Recruiter dashboard with advanced features
• Candidate portal
• Email/SMS notifications
• Analytics and reporting

• Machine learning models
• Multi-language support
• Mobile application
• ATS integrations

See PROJECT_REPORT.md for complete roadmap.

• Time per resume: 5-10 minutes
• Cost per resume: $2.50-8.33

• Processing time: 10 seconds
• Cost per resume: ~$0.01

Savings: 99% cost reduction | $30,000-100,000 annually for 1000 resumes/month

• Anshuman Mohapatra - Team Lead, Backend Pipeline & APIs
• Shivam - Skill Parsing & Data Extraction
• Siddhant - Job Matching Algorithm
• Pavan - Frontend Dashboard UI

Mentor: Kabir

• PROJECT_REPORT.md - Comprehensive project documentation
• docs/handoff-notes/ - Technical handoff documentation

# Hard refresh browser
Cmd+Shift+R (Mac) or Ctrl+Shift+R (Windows)

• Check internet connection
• Verify AWS credentials
• Check CloudWatch logs
• Ensure S3 bucket exists

• Refresh the page
• Check DynamoDB tables
• Verify API Gateway endpoints
• Check browser console for errors

This project is part of an academic assignment.

• AWS for providing cloud infrastructure
• AWS Textract for AI-powered text extraction
• Our mentor Kabir for guidance and support

For questions or support, please contact the team members through GitHub.

If you found this project helpful, please give it a ⭐️!

Built with ❤️ using AWS Serverless Architecture