Chomp

Inspiration

2 years ago, one of our founders (Riddhiman) was diagnosed with diabetes with a sugar level of 5.8%, just on the border of type 1 diabetes.

What it does

This app was designed in order to help him decide what foods to eat. Before a person eats something, they can scan the food's nutrition label on the back and be able to more easily decide for themselves if eating something is a good idea.

How we built it

We built it using the t3 stack. We used sqlite as a database and used pytesserat for the image text recognition.

Challenges we ran into

We often found it difficult to merge each of our code bases together. One person would have a completely different set of files and it took a while and a lot of effort to merge it together.

Accomplishments that we're proud of

Making an almost functional app with a nice looking ui despite the time constraint.

What we learned

• AI Image recognition
• How to deal with uploading files
• Why typesafety is not the best idea during a hackathon

What's next for Chomp

• Smart Glucose Tracker connectivity, to directly inference real-time sugar data, and make better AI predictions
• Change of database
• More authentication options and features

Built With

• next.js
• typescript