-
-
Home Page
-
Home Page (Scrolled Down)
-
Challenge Page (Not Logged In)
-
Challenge Page (Logged In)
-
Challenge Creation Page
-
Leaderboard Page
Inspiration: We were inspired to make this project because our team loves exploring campus and wanted to provide the UT community with a fun web app similar to a scavenger hunt while also allowing them to discover and learn about new areas of campus.
What it does: Our project is a web app that allows users to post pictures taken from different locations on the UT campus. In the submission process, the user’s physical geolocation is recorded. Other users can then guess the location where the picture was taken, be physically present in the guessing location, upload a picture of the same place at similar angles, and the web app will then check both the physical location and image similarities to determine the user’s success. Once successful, 10 honks will be rewarded to the user. The web app has a leaderboard feature for the users to compete with each other for the number of honks.
How we built it: We used Javascript, Typescript Language, and Cursor code editor to help us code the front and backend of the web app. Deployed the backend on Firebase. We used npm to build the content, and a remote VPS with Nginx installed to deploy the web server with a domain name.
Challenges we ran into: Ran into a Firebase issue where we had issues setting it up GitHub: had to split into multiple branches of the same repository because we kept breaking the code on the main branch.
Having git commit and git pull problem. GPS coordinates were wildly inaccurate Accomplishments that we're proud of: We got the GPS coordinates working (to within 50 meters). We got the Machine Vision model (GPT 5.2) working and integrated it into our project’s backend to compare images
What we learned: How to setup FireBase How to use Cursor to help with some parts of the programming How to collaborate and stay organized under tight deadlines
Log in or sign up for Devpost to join the conversation.