FreshFridge

Inspiration

Our inspiration for this project was the issue of food waste in our community. There's a lot of food waste in our city, but not many people think of giving away or selling this food for budget prices to serve the community as well as decrease food waste. We did some research on this issue, and discovered that worldwide, 1.3 billion tons of food are wasted every year! Thus, we created FreshFridge.

How we built it

We used a lot of different technologies to build this app. In the frontend, we used Next.js as our main driver of the user interface, as well as Radix UI and Tailwind CSS to make it easy for us to import beautiful components like forms and textboxes into our application. We used Clerk as an authenticator, ensuring safety by checking whether the user is signed in or not. In the backend, we used Quart.py to handle routing, the OpenAI API to generate recipes based on food input, a TensorFlow model to analyze the picture of the users' fridge, and Airtable to manage and receive data.

Challenges we ran into

A major challenge that we ran into was making the design of the marketplace, as we needed to ensure that it could display many products while still being simple. We decided to opt for a grid of "cards", or small informational rectangles on each one of our products, similar to browsing through YouTube or Amazon. We also decided to make a separate page for checking out items from the marketplace. Our smart design is similar to Gmail, where we have a list of checked-out items and ways to interact with these items (such as removing them from the cart).

Accomplishments that we're proud of

One accomplishment that we're proud of is using a new UI library, Radix UI. It was quite different than what we were used to, but it was a pretty big upgrade and its components were simple and easy to modify. Another accomplishment that we're proud of is successfully making a very fast TensorFlow model to process our images and process them with our OpenAI API interactions.

What we learned

The main thing that we learned from this hackathon is time management. We were able to successfully budget time for different tasks, which made our work a lot more efficient. By budgeting time for certain tasks, we could ensure that we would have the app done long before the deadline, which would leave enough time for presentation building and any debugging.

What's next for FreshFridge

One idea that we have that could further improve FreshFridge is more accurate expiration date detection for images of users' fridges. This would improve our app by making sure that food is likely safe to eat. We could also implement methods of detecting the store prices of certain products based on barcodes, so people could re-use recipes that we generate by purchasing products at a grocery store.