Inspiration
The rising grocery prices in Canada suggest that buyers are seeking ways to save what they can by cutting their costs, purchasing alternatives, and occasionally skipping meals. The struggles of the people are deepening as groceries become more and more inaccessible and food insecurity grows.
What it does
CuisiNet is an AI powered tool to be used to help users find their blind spots in the search for affordable sustenance by suggesting locally inexpensive groceries to add to their shopping cart every week to optimize their savings from weekly expenses.
How we built it
Google's Gemini was used to create a majority of boilerplate code, implementing the design of the user interface and parsing the responses from the Gemini API. The boilerplate code was then adjusted to add the developers' customization in creating the application.
Challenges we ran into
The AI models presented were all very appealing, and made it very difficult for us to decide which model to use for our purposes. After consideration, it was decided that since Google's AI model allows for corrections to ensure that the data provided is up to date.
Gemini's studio, however, had slowly veered off track and resulted in slowly "forgetting" previously implemented functionality. It was necessary to implement features slowly and thoroughly while also debugging all functionality after each change to ensure features were implemented properly.
Accomplishments that we're proud of
This was the first time, for both of us (group of 2), to have created an application using AI generated code. I would prefer to have created an application on our own as we did in a previous hackathon, however, the project would have been very difficult to even begin, especially without any knowledge of how to use AI model API response parsing.
What we learned
The use of AI can be used to create boilerplate code to ensure that mistakes can generally be avoided, especially for my partner and I who struggle with UI and UX design and would prefer to implement our ideas without being stuck on visual design and focus on the features of our apps.
What's next for CuisiNet
We're thinking of adding further features to help with food accessibility, including recipes which can be considered when finding new groceries to purchase, with more of a goal to help introduce new foods to a public who is suffering from food insecurity.
Log in or sign up for Devpost to join the conversation.