jobGPT

Inspiration

I wanted to build a job search application that utilizes AI to find the most relevant job listings based on a user's query.

What it does

The application takes a user's query, along with additional filters such as the companies and the number of job listings to display, and returns the most relevant job listings. It achieves this by using semantic search to match the user's query with available job listings in the database.

How I built it

The application is built using Python, FastAPI, OpenAI, and Postgres. The PostgreSQL database is used to store job information and there is also a custom scraper I wrote to get many job listings which I then tried embedding using various different embedding models to get high quality matches.

Challenges we ran into

1. Generating high quality embeddings given some longer job listings.
2. Scraping job data from different sources and storing it in a PostgreSQL database.
3. Integrating the ChatGPTPlugin with the FastAPI application.

Accomplishments that we're proud of

1. Successfully building a functioning job search application using AI.
2. Creating a well-structured application with organized code.
3. Implementing a user-friendly API that allows users to easily search for relevant job listings.

What we learned

1. How to use FastAPI to build a web application.
2. How to scrape job listings.
3. How to work with PostgreSQL databases in Python.

What's next for jobGPT

1. Expand the job sources to include more job listings and improve search results.
2. Implement a more advanced filtering system to allow users to find the most relevant job listings based on their preferences.
3. Improve the user interface and make the application more accessible on various devices.

Built With

• fastapi
• numpy
• openai
• postgresql
• python