I built ai-job-application-copilot, a Python application that analyzes resumes against job descriptions, identifies matching skills and potential skill gaps, and generates professional job application messages. My GitHub profile lists the repository among my projects: github profile.
- The application extracts skills from resume text, identifies skills mentioned in job requirements, highlights skills detected in both, and flags job requirements not detected in the resume. Its message generator uses detected matching skills to create a professional, role-specific application message. Input validation rejects empty and excessively long inputs; automated tests cover API endpoints, validation, skill analysis, and message generation.
I use Python, FastAPI, Pydantic, and Pytest. The project is organized around app/main.py, app/analyzer.py, and app/message_generator.py, with tests in tests/test_main.py and tests/test_message_generator.py. It also includes requirements.txt, ai-usage-log.md, intent.md, and README.md.
- The API provides GET / for application status, GET /health for a health check, POST /analyze for resume-skill analysis against job requirements, and POST /generate-message for generating an application message. The POST endpoints accept a resume and job description as text.
Skill detection relies on a predefined skill catalog and phrase matching, so a potential gap means only that a skill was not detected in the supplied resume; it does not establish that the candidate lacks that skill. Generated messages are template-based drafts that should be reviewed before sending. The project does not make hiring decisions or guarantee job suitability.
- The README lists semantic skill matching, PDF and DOCX resume uploads, customizable message templates, NLP- or LLM-based skill extraction, and a web interface as future improvements.
My README reports 9 passing tests in the current test suite.
python -m venv .venv
pip install -r requirements.txt
uvicorn app.main:app --reload
python -m pytest -q