Inspiration
I was inspired by what I consider "fun". I love twisting regular elements. In this case, I used an API that is designed to return current job openings, and turned it into clustered data for Programmatic SEO.
What it does
It generates human-readable, aggregated content from job posts on CrackedDevs. I thought it would be useful if I wrote some code that may inspire the platform itself, instead of being forgotten after the Hackathon ends.
How we built it
I got the idea and spent my night coding it on Jupyter Notebook. That's the entire story.
Challenges we ran into
The CrackedDevs API data was curated for another task, which is listing the job posts. Converting the data into the desired format using LLMs was not hard, but required some thinking.
I also had to optimize my prompts in a limited time for better accuracy, but I could overcome this challenge easily.
Accomplishments that we're proud of
The semantic clusterization part. That's the entire core of the project.
What we learned
That all the data we need are just in front of us. Manipulating into the right format is what we all need.
What's next for Programmatic SEO using CrackedDevs API and LLMs
Better templates and a good design. I may design some pages with graphs and infographics so that the content would be better.
Built With
- jupyter
- llms
- python
Log in or sign up for Devpost to join the conversation.