We wanted to build an app that would help people make smarter grocery shopping choices. With the increasing challenge of food waste, it's easy for people to forget that they can make smarter choices and save the planet.

What it does

The mobile app lets users scan food items and create an inventory of items. Once the app has learnt a user's shopping habits, it makes recommendations on what to reduce based on the frequency of purchase and use. Using machine learning and Google Vision image recognition software, the app detects the food item and makes a list of food in the inventory along with their quantities. A user can update their usage by editing an item and wastage details by deleting an item.

How we built it

We built all the app frontend using HTML, CSS and Javascript. The data was stored in Cockroach DB's servers and the Google Vision API.

Challenges we ran into

Half the team were beginners to hackathons and none of us had used Cockroach DB or Google Vision API, so figuring it out within the time limit was a challenge. We had initially chosen Expo as the tech stack for the frontend, but couldn't figure out React native in time to use it so we changed gears halfway through and used HTML, CSS and Javascript to code the frontend.

Accomplishments that we're proud of

We're proud of completing a working project and learning to use CockroachDB.

What we learned

We learnt the basics of image recognition software and the software development life cycle. We also realized the importance of communication and figured out how to collaborate virtually.

What's next for Smart Cart

In the future, we'd like to track better-informed grocery recommendations and recommended recipes based on the items available. We'd also like to give users a warning if their wastage is higher than their usage.

Built With

Share this project: