Resume Screener

*** ⚠️ This project is being completely reworked and continued by Jahleel ⚠️ ***

Quick Note

• This project was originally a group project, which can be found here, but because I have completely overhauled the backend (and currently working on the frontend), their names have been removed. The original proof-of-concept project will remain at the aforementioned link.

Current Version

• Frontend:

  • TypeScript: Enforces type-safety and stronger development than plain JS
  • React: Creates a rich and interactive UI for an enhanced UX.
  • Flask-CORS: Just for routing

• Backend:

  • Python: Base language for the backend to handle endpoints, JWT auth, and orchestration of AI workflows, as well as both database interactions.
  • Flask: Framework used to build the API.
  • LangChain: Used to orchestrate OpenAI with various APIs
  • OpenAI: Cloud-based API to access GPT 4o Mini
  • Vector Database (Pinecone): Used for embedding the unstructured data (resume and job descriptions), enabling efficient similarity search and retrieval, allowing the LLM to analyze and compare content more efficiently.
  • SQLAlchemy: Pythonic SQL for object relational mapping (between User, Resume, and Analysis types)
  • Flask-JWT-Extended: Managing JWT auth

Former Version

• An AI-powered application that compares your resume against a job posting and provides actionable feedback. Built with Flask, MongoDB, Playwright, and OpenAI's API. All logic now lives in a unified backend structure.

Table of Contents

• Resume Screener
  • Quick Note
  • Current Version
  • Former Version
  • Table of Contents
  • Prerequisites
  • Environment Variables
  • Setup & Run Locally
  • Docker Compose
  • Running Tests
• For py
  • Container Images
  • Deployment
  • License

Prerequisites

• Docker & Docker Compose
• Python 3.10+
• Node.js 18.20+
• PostgreSQL
• OpenAI API
• Pinecone API

Environment Variables

1. Create a .env file in the directory based on the provided .env.example.
2. Include the following variables:

MONGO_URI = mongodb://mongo:27017/resume_db
SECRET_KEY = your_secret_key
OPENAI_API_KEY = sk-xxxxxxxxxxxxxxxxxxxxxxxxxxx

1. After copying, rename to .env and replace dummy values, including the OpenAI Key, with your own.

Setup & Run Locally

1. Clone the repository:

   git clone https://github.com/JahleelT/jahleelt-resume-screener.git
   cd jahleelt-resume-screener

2. Copy env.example to .env in , then update values.
3. Build and start services with Docker Compose:

   docker-compose up --build

4. Visit http://localhost:5050 to access the Web

Docker Compose

The docker-compose.yml file defines:

• mongo: MongoDB service on port 27017
• app: Unified on port 5050

Running Tests

*** ⚠️ Tests are currently on hold until further notice as I migrate the repository from two subsystems to one unified system ⚠️ ***

# For py
cd app
pytest --disable-warnings -q


## Container Images

- [gbs4189/web-app](https://hub.docker.com/r/gbs4189/web-app)
- [gbs4189/ml-client](https://hub.docker.com/r/gbs4189/ml-client)

## Deployment

We deploy to DigitalOcean via GitHub Actions. Configuration is defined in `spec.yaml`. Ensure your DigitalOcean App ID and required secrets are set in GitHub repository settings under **Secrets and variables**.

## License

This project is licensed under the MIT License. See [LICENSE](LICENSE) for details.