webInternal ProjectFeatured
Resume Scoring & Candidate Ranking System
Web application with semantic resume analysis using RAG and LLMs. Complete UI/UX design and full-stack development.
Developed at ATP
Full Stack Developer
2024
Internal project - no public access
Screenshots
About This Project
Built a semantic resume analysis system that uses RAG (Retrieval Augmented Generation) and LLMs to automatically score and rank candidates based on job descriptions. The system extracts key information from resumes, matches skills and experience against job requirements, and provides recruiters with ranked candidate lists and detailed match explanations.
Key Features
- Semantic resume parsing and information extraction
- RAG-based matching against job descriptions
- Automatic candidate scoring and ranking
- Detailed match explanations for each candidate
- Batch processing for large applicant pools
- Export functionality for ATS integration
Challenges & Solutions
- Handling diverse resume formats (PDF, DOCX, images)
- Building accurate semantic matching beyond keyword matching
- Scaling to process thousands of resumes efficiently
- Reducing bias in AI-powered candidate evaluation
Results & Impact
- 40% reduction in resume screening time
- Improved candidate quality in shortlists
- Higher recruiter satisfaction scores
Technologies Used
Next.jsTypeScriptPythonOpenAIPineconeFastAPITailwind CSS
Project Details
- Category
- web
- My Role
- Full Stack Developer
- Duration
- 4 months
- Year
- 2024
- Company
- ATP
- Status
- Internal / Private
Internal Project
This project was developed for internal use at ATP. Source code and live demo are not publicly available due to confidentiality.