FRIDGY

Landing Page
Features
Dashboard 1
Dashboard 2

Inspiration

We often found ourselves forgetting what we needed during frequent trips to Costco and Shoprite. That’s why we created FRIDGY—to simplify life by tracking your fridge’s capacity, ingredients, and when it's time to restock!

What it does

FRIDGY analyzes photos you upload, using Gemini Vision to identify ingredients and calculate fridge capacity. It converts the ingredient data into JSON and stores it in MongoDB, while the photos are saved locally. An algorithm predicts when you’ll run out of food using a slope formula, and it integrates with Google Calendar to schedule shopping trips based on your current events and even the weather!

How we built it

We used Python, HTML, CSS, and native JavaScript. We also created a custom package for common functions called py_tools, and utilized libraries like Flask.

Challenges we ran into

Google Calendar authentication was the toughest part—it threw errors and took over two days to resolve.

Accomplishments that we're proud of

This was our first time using MongoDB, and tackling such a complex project taught us how to organize large files into smaller, manageable components.

What we learned

Clean code is key! Going forward, we'll limit files to 2-3 functions max and ensure tidier code, especially for messy files like JSON_EDITOR, which was hard to debug.