NutriSnap
Point your phone at your meal tray, get macros instantly. Built in 24 hours at Cheesehacks 2026.
Overview
NutriSnap started as a hackathon idea: what if UW-Madison students could photograph their dining hall tray and get accurate nutrition info without guessing? We built and deployed a full-stack nutrition tracker for UW–Madison dining halls at CheeseHacks 2026 in a 4-person team — frontend on Cloudflare Pages, REST API on Render. We integrated Google Gemini 2.0 Flash for image-based food recognition across 6 dining halls, pulling live menu data from the Nutrislice API to match recognized items. The backend is a JWT-authenticated REST API with BMR-based macro targeting and a PostgreSQL-backed meal logging system with a 30-day macro history dashboard.
What it does
Google Gemini 2.0 Flash for image-based food recognition across 6 dining halls
Live menu data via Nutrislice API — no hardcoded menus
JWT-authenticated REST API with BMR-based macro targeting
PostgreSQL-backed meal logging with 30-day macro history dashboard
Frontend on Cloudflare Pages, REST API on Render
Built by a team of 4 in ~24 hours at CheeseHacks 2026
Built with