AI-powered Resume Matcher using LLaMA 3 + Streamlit
100% local, private, recruiter-ready. Built with Python, Ollama, and open-source tools.
ResumeMatcherAI allows recruiters or hiring managers to:
- Upload a Job Description (JD)
- Upload one or more candidate resumes
- Get an instant score out of 100 + explanation
- Download results as CSV or Excel
All powered locally via Ollama + LLaMA 3, ensuring privacy and speed — no cloud required.
- 🐍 Python
- 🧠 Ollama (LLaMA 3)
- 🌐 Streamlit
- 📄 PDF / DOCX parser
- 🗂 Notion / Airtable-ready output (optional)
- Upload JD (.txt) + Resumes (.pdf, .docx)
- Score resumes using local LLM
- View results in a browser UI
- Download CSV/Excel with Score + Reason
- Fully offline-capable
- Extensible: Notion, Airtable, Email
Replace this with a screenshot or demo GIF of your Streamlit UI
ResumeMatcherAI/
├── input/
│ ├── jd.txt
│ └── resumes/
├── output/
│ ├── scores.csv / scores.xlsx
├── web_app.py # Streamlit UI
├── app.py # Backend logic
├── prompt.txt # LLM scoring prompt
├── README.md
- Clone the repository
git clone https://github.com/aneelv75/ResumeMatcherAI.git
cd ResumeMatcherAI- Install dependencies
pip install -r requirements.txt- Install & run Ollama + LLaMA 3
brew install ollama
ollama run llama3- Launch the web app
streamlit run web_app.py| Resume | Score | Reason |
|---|---|---|
| john_doe.pdf | 86 | Matches Python + NLP, lacks domain exposure |
| jane_smith.docx | 65 | Good experience but not JD-aligned |
- Add multi-JD matching
- Auto-email shortlists to hiring managers
- Integrate Notion & Airtable dashboards
- Schedule auto-runs (daily/weekly)
Built with ❤️ by Anil V
This project is licensed under the MIT License
