Recipez

Homescreen (generate recipes)
Fridge
Saved Recipes
Profile

Inspiration

After a long day of work, figuring out what to cook can take a lot of time and energy; both of which college students lack. When at home, it can be difficult to know what recipes to make with limited ingredients. When at the grocery store, it can be difficult to know what ingredients you are missing at home for certain recipes. Our product aims to solve both of these problems, acting as an automated recipe and ingredient manager.

What it does

Our product keeps track of all the ingredients in your home by allowing you to scan receipts from the grocery store. Then, based on the ingredients available, our product is able to suggest recipes you would be able to make. If you find a recipe you like, you can save it, and the ingredients used will be removed whenever you make it.

How we built it

We utilized react-native Expo and leveraged Google's Cloud Vision API for OCR, Gemini to generate recipes, and Firebase for the backend.

Challenges we ran into

We faced a few key challenges, especially around OCR accuracy and text parsing from receipts. Cleaning up inconsistent item formatting (e.g., brand names, typos, and different units) required a lot of trial and error with prompt engineering.

Accomplishments that we're proud of

We successfully integrated image-based item detection using OCR and natural language processing, turning a simple receipt image into structured food data.

What we learned

We learned how to utilize API architecture effectively, especially chaining two different services (Google Vision + Gemini) to create a smooth user experience. We also deepened our understanding of React Native’s component lifecycle, image handling, and state management.