About

This repo uses Cohere's embedding API and Streamlit to build a job recommendation based on user's input resume. The database used is Cohere's job postings as of 1/24/2024. Postings are scraped, parsed and stored in this repo.

Environment Setup

python3 -m venv venv/
pip install -r requirements.txt

Dataset Setup

Run the job scraper:

python3 scraper/job_scraper.py

Next, run the description parser

python3 job_html_description_parser.py

Running App

CO_API_KEY={REDACTED} CO_MODEL_NAME="embed-english-v2.0" streamlit run main.py

Future Improvements

• Only supports text-based resumes. Run your PDF resume through an online PDF to text converter and then make touch-ups as needed.
• Filtering based on other criteria (location, remote, etc.)