Applicant tracking system

This repository contains a resume parsing and matching project. It extracts text from resumes, parses candidate information, matches resumes to job descriptions, and generates explainable AI (XAI) reports. The project appears to include multiple entry points and utilities for parsing, indexing, matching and producing reports.

Key features

• Extract text from resume files (PDF/DOCX/TXT).
• Parse structured information (name, contact, education, skills, experience) into JSON (parsed_resumes.json).
• Match parsed resumes to job descriptions and produce ranked results.json.
• Generate XAI-style explanation reports for candidate rankings (xai_report.py / reports/).
• Small web/app entrypoints (possible app.py / main.py) for quick demos.

Repository structure (important files)

• app.py — possible web or demo app (start point may vary).
• main.py — another entrypoint; check which one you use to run the pipeline.
• parse_resumes_gemini.py — resume parsing using a model or LLM-based approach.
• extract_text.py — resume text extraction utilities.
• resume_matcher.py — matching logic to compare resumes with job descriptions.
• xai_report.py — produces explainability reports (XAI) stored under reports/.
• parsed_resumes.json — sample or output parsed resumes.
• results.json — sample or output matching results.
• data/ — contains job descriptions (data/jd/) and resume source files (data/resume_data/).
• reports/ — generated XAI reports for candidates.