Inspiration

Collaboration is heralded as a pillar for higher learning. Institutions encourage, if not almost require it, for students to connect with each other and help one another learn. However, there are three fundamental challenges with the current paradigm for finding peers to collaborate with:

  1. Many students lack knowledge of a list of students in their class looking for partners
  2. Many students lack access to information about the academic background of partners
  3. Many students lack the social confidence to ask a peer to partner with them With these challenges in mind, Firestorm provides an interactive web platform to make finding problem set partners seamlessly.

What it does

Students create profiles by registering in the site, and are able to view profiles of other students whom they can choose to approve or reject to be pset partners. If two students both approve each other, they are given each others' Kerberos accounts to contact.

How we built it

The frontend of the website was built as a React app and used Firebase to manage backend. To recommend potential partners, we programmed an ELO system where potential partners with similar ELOs are recommended to one another.

Challenges we ran into

We faced challenges with implementing the navigational tools in the website properly, where certain buttons would not route to the correct places. Additionally, we faced challenges with file uploads of profile pictures, having trouble saving them correctly and displaying on the main page. As a team of only two, we faced limitations on time management as well.

Accomplishments that we're proud of

Building a functional web platform frontend and backend, while implementing our own recommendation algorithm.

What we learned

Had limited experience with backend development prior to the hackathon, but gained familiarity with Firebase services. Improved frontend skills.

What's next for Firestorm

Improving the UI, incorporating more robust matching algorithms (perhaps with the help of ML or AI), and adding more features for profiles.

Share this project:

Updates