CourseAI (Productivity & Automation)

Inspiration

As UMD students, we've all suffered the many hours of sorting through Testudo's schedule of classes, trying to figure out what GenEds we needed while getting our major requirements. We wanted to develop an assistant that could parse through the seemingly endless amount of courses and details, and get the information straight to the student easily.

What it does

CourseAI uses Google Gemini to scan through all the course information listed on Testudo and return the courses specific to the user query, so that any student could ask the simple question of, "which class offers XYZ gen eds without ABC prereq," and immediately get results.

How we built it

We used a web scraper for the Testudo schedule of classes, converting all the relevant information into a JSON file that could be scanned by Google Gemini's API, and then have the relevant sections of said JSON file displayed to the user.

Challenges we ran into

We had many issues with compilation and configuring the webscraper and API to be able to interact with each other.

Accomplishments that we're proud of

Having the flexibility of being able to ask generic and specific queries. A student could ask for any 3 credit classes, for instance, and be given all the results, or if they needed something specific to fit a circumstance, they could simply explain the situation in their query, and still have the program output exactly what they need. This was the first time either of us had ever worked with Google Gemini API, and we found the lack of monetary spending to be great! We've both had misfortune using other AI API's in the past

What we learned

Our main goal was to get better at utilizing API's and scraping, and this project was perfect for both.

What's next for CourseAI

Course requirements, credits, and professors are all good, however, more data is always beneficial for an app like this. Teacher reviews and class stats would be incredibly helpful for students using this project.