NVIDIA 2026 Hackathon Winner

Built a computer-vision triage tool that helps cities prioritize road repairs by ranking defects by the cost of leaving them unresolved. Built in three days, it won first place.

That was at NVIDIA Spark Hack Toronto 2026, judged on the three days of work you can open above.

Three days, and one hard rule: public APIs only, nothing private, nothing behind a contract. A city already knows where its roads are failing. The reports exist. They just arrive as a flat queue with no way to tell which crack becomes a pothole in six weeks and which becomes a claim. The problem was not collecting more data. It was turning fragmented reports into a real-time coordination platform where residents and dispatchers could keep the city updated, prioritize issues, and resolve them together.

We trained our own crack-detection model on NVIDIA hardware, a desktop-sized box on the table beside us, using NVIDIA NeMo, their open-source toolkit for training and customising AI models, then hosted it on Hugging Face so the app could call it live. Around that I designed one screen that serves two very different people: a virtual twin of Toronto, the real street grid rebuilt in 3D from the city’s own open data, that any resident can read and fly through, and a ranked dispatcher queue that puts a dollar-per-day cost of delay against every ticket. My role was to bridge the model and the product, shaping the underlying technology into an experience people could actually use.