Rescan

An AI-powered Applicant Tracking System that analyzes resumes against job descriptions to find the best matches using large language models.

Features

• PDF extraction and processing for both resumes and job descriptions.
• Advanced document analysis using Ollama LLMs
• Skills matching and scoring
• Education and experience validation
• Improvement suggestions for candidates
• Clean API design with FastAPI
• Containerized deployment with Docker

Tech Stack

Server: FastAPI

AI/ML Ollama models(llama3.2)

Package manager uv

Testing pytest

Prerequisites

• Python 3.12+
• Ollama running locally or accessible via network
• Docker (optional, for containerized deployment)

Run Locally

Manual Setup

Clone the project

  git clone https://github.com/Csb-218/Rescan.git

Go to the project directory

  cd rescan

   uv venv .venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate

Install dependencies

  uv sync --locked --no-install-project

Start the server

  fastapi run 

Docker Setup

Build image

docker build -t rescan .

Run the container

docker run -p 8000:8000 rescan

Environment Variables

To run this project, you will need to add the following environment variables to your .env file

OLLAMA_BASE_URL

API Endpoints

/analyze/advanced Analyzes a resume against a job description using advanced LLM matching.

Method: POST

Parameters:

jd: Job description file (PDF)

resume: Resume file (PDF)

Returns: JSON with match details, skills comparison, and suggestions

Usage/Examples

1. Start the application

2. Send a POST request to /analyze/advanced with both resume and job description PDFs.

3. Receive detailed analysis including:

   -> Skills matching

   -> Experience validation

   -> Education requirements check

   -> Overall match score

   -> Improvement suggestions

Screenshots

Running Tests

To run tests, run the following command

  pytest 

License

MIT

Contributing

Contributions are always welcome!

Please feel free to submit a Pull Request.