SmartCart

Inspiration

What it does# SmartCart

Inspiration

We wanted to make grocery shopping smarter, healthier, and more sustainable. With so many options, confusing labels, and rising food costs, shoppers often struggle to make the best choices. Our inspiration was to create a tool that empowers people to eat better, save money, and reduce food waste all in one seamless experience.

What it does

SmartCart lets users create a health profile, scan products for a health score analysis using our custom regression model, generate a health score analysis for each product, integrate retailer carts, track spending, offer nutrition coaching (Lumi), manage pantry tracking, and suggest smart meals.

How we built it

The frontend was developed with SwiftUI in Xcode, and the database is powered by Google Firebase. Python was used for Flask servers to host the trained models and handle secure API calls.

Challenges

Balancing so many features within a short hackathon timeframe.
Integrating different systems (Google A2A, scoring model, ensuring seamless server processes) into a single flow.
Designing an AI assistant that feels friendly, credible, and approachable.
Managing the trade-off between technical complexity and a clear, demo-ready prototype.

Accomplishments

We built a working prototype, created a clear user journey, designed a strong brand identity, and collaborated effectively all within 36 hours.

What we learned

Teamwork is everything. Tasks should be outlined and divided for smoother progress. We also realized this is just a hackathon and we’re not deploying an app tomorrow. Sometimes mock data is the best way to go for non-core features.