FridgeSync

Landing Page
Dashboard
Fridge Stats
Recipe Generator
Generated Recipe

Inspiration

One of our teammate's grandma often has issues with managing her fridge and pantry—it gets hard to keep track of what she has and doesn’t have when thinking of what to make for dinner or when going grocery shopping. We aimed to create an app that would help her, and others who have similar issues or simply want more convenience in their lives. That’s where FridgeSync comes in.

What it does

FridgeSync allows users to easily scan and store items and ingredients bought from the store, and keep track of them in a simple interface. By taking a photo of an item, it will be added into your virtual fridge, and information about the item will be generated. From this virtual fridge, users can generate nutritional insights, recipes, grocery lists, and more.

How we built it

We leveraged Svelte with Tailwind CSS to create our frontend, and Svelte API routes to power our backend. In our backend, we used Supabase (a PostgreSQL cloud database) to keep track of the items and photos, OpenAI’s Vision API to handle image recognition, and OpenAI’s GPT-4 API to generate recipes and grocery lists.

Challenges we ran into

We took quite a bit of time to initially come up with our idea. We faced numerous bugs while trying new libraries and faced a lot of styling issues as well. Furthermore, we had to repeatedly prompt engineer our API requests so that we could be given an accurate response to display to the user.

Accomplishments that we're proud of

We're proud of how fast we created new features once we got our idea down. Additionally, we're proud that UI/UX has improved dramatically compared to past projects.