Park Genius was born out of a real frustration. We weren’t even planning to compete in the hackathon — we had just come to do henna tattoos — but after circling full parking lots at Mason and wasting gas, we realized how inefficient the system felt. That frustration sparked an idea: what if students could see the closest available lot before arriving and reduce congestion by carpooling with classmates who share similar schedules?
We built Park Genius using Python with Flask for the backend, structuring our system around APIs, geolocation services, and modular data flows. Students can view a live campus map showing the nearest lots and estimated available spots. Since universities don’t provide public real-time parking APIs, we implemented a crowdsourced check-in system where students log when they park or leave, continuously updating availability. We also trained a machine learning model using scikit-learn to analyze parking patterns and predict peak congestion times, allowing the app to suggest smarter parking options. The front end was built with HTML, CSS, and JavaScript, and in total, the project spans over 5,000 lines of code. We also gamified the experience — students earn achievements and incentives for checking in — encouraging consistent participation and improving data accuracy.
One of our biggest challenges was designing a reliable prediction system without official real-time datasets, so we had to engineer a scalable, data-driven model from the ground up. We’re proud that Park Genius isn’t just a concept — it’s a working system that reduces wasted gas, saves time, and promotes sustainability. Even better, it’s built to be adaptable and ready for deployment on any commuter-heavy campus, not just George Mason.
Log in or sign up for Devpost to join the conversation.