Inspiration

We are all seniors, so we decided to have fun for this last hackathon and build a goofy project together. Initially we were just planning on attending workshops and not really building anything too big, but after a laughter filled brainstorming session, we decided this was too good to pass up.

What it does

The Performance Evaluation Engine\(^{TM}\) tracks developers based on their two most important statistics: GitHub commit history and Clash Royale records. Thanks to the generous APIs that both companies provide, we are able to pull the data that matters most to employers. The Engine provides everyone in the group with a chart that shows them how they stack up against everyone else. If you're falling behind, you might end up in the danger zone! Be sure to get out of the danger zone fast using our deck tutor, or else you'll be in trouble.

How we built it

We used Next.js and React to build out a frontend and an API, and we used Atlas MongoDB for the database and Google Gemini for our deck tutor tool.

Challenges we ran into

One of the biggest challenges we ran into was unintentionally stepping on each other's toes. There were a number of occasions where two developers would be working on similar tasks, and while solving their own task, would begin to develop overlapping features with others. This led to confusion later on when other developers wanted to use duplicate functionality and didn't know which to choose. We also ran into the issue of burnout. Despite all being dedicated and working through the night, lacking sleep catches up with you. Because of this we slowed down substantially as the night went on.

Accomplishments that we're proud of

We're really proud of our Deck Tutor feature, which gets match history for a user in Clash Royale in near real time, parses and cleans it, and sends it to Google Gemini for recommendations and lessons on how to improve. We're also really proud of the GUI. Due to the very clash royale heavy nature of our project, we took a lot of inspiration from the game's look and feel, and we believe it comes across as a genuine - if not very early beta - Supercell product.

What we learned

We learned that when you have a dedicated team of high skill software engineers, the development process can fly by. Not having to learn git or teach a team member how to run a development server saves so much time and headache as opposed to the tedious beginner process. We also learned that relying on generative AI, even for relatively small projects, leads to heavy and rapid buildup of technical debt. We avoided using AI until the end - when we started to hit crunch time - and we still quickly found ourselves lost in the Claude slop.

What's next for Performance Evaluation Engine

While it was super fun to build this project as a group of friends about to graduate, we don't expect to revisit this much beyond this hackathon, except maybe to show to friends and make them laugh. Ultimately, this was a last hoorah in which doing whatever we wanted was our first priority.

Built With

Share this project:

Updates