Inspiration

Navigating through Testudo and ELMS can often be a daunting task, particularly during class registration. Students find themselves juggling multiple tabs to compare class times, seat availability, professor ratings, and campus maps to ensure they can make it to class on time. Recognizing this challenge, we’ve developed a user-friendly, all-in-one application tailored specifically for UMD students. This innovative platform streamlines the process, making it a seamless and stress-free experience.

What it does

TerPilot is a cutting-edge platform that seamlessly integrates traditional algorithms with Large Language Models (LLMs) to revolutionize course selection, schedule generation, and academic progress tracking. By tackling the NP-hard problem of schedule planning, TerPilot ensures that students can effortlessly navigate their academic journey. Its sophisticated algorithms work with the user-friendly interface, providing an unparalleled experience for students aiming to graduate successfully.

How we built it

The frontend was developed using React for a dynamic and responsive user interface. For the backend, we used Flask with Python, and PostgreSQL for the database. We leveraged scraped data from Testudo and PlanetTerp APIs with the help of BeautifulSoup to obtain relevant information for our algorithms and LLM.

Challenges we ran into

One of the most time-consuming tasks was to process the vast amount of scraped data. We ended up with over 14 million characters of data that we store in our database.

Accomplishments that we're proud of

Integrating multiple tools to create a large project in 36 hours with a team of three people.

What we learned

We challenged ourselves to use new tools and learn techniques we had never used before or feared using. However, we gained valuable knowledge that will aid us in future hackathons.

What's next for TerPilot

We plan to integrate terpilot to ELMS and gradescope for direct access to grades and schedules. Furthermore, we plan to give it access to latest news and events on-campus. Finally, we aim to improve our algorithms and increase the efficiency of our LLMs to be more sustainable.

Share this project:

Updates