ShopSmart

Inspiration

Prior to ShopSmart, our team was inspired by the growing issue of food waste and the detrimental environmental consequences that come with it. According to the United States Department of Agriculture (USDA), over one-third of all available food goes uneaten through loss or waste. Consequently, this significantly contributes to greenhouse gas emissions, and worldwide food loss is a pervasive global issue. Meanwhile, many families face rising grocery costs and limited access to sustainable food options without direct opportunities to mitigate these challenges. From this, we wanted to create a solution that empowers individuals to reduce waste at home while shifting towards sustainability.

What it does

ShopSmart's primary aim is to democratize food accessibility whilst minimizing food waste worldwide. Through our product, we help create custom recipe suggestions based on both local price and nearest location. Firstly, we have a "Shopping List" tab that allows the user to add any item to their grocery list, which provides the location of where to find the item and the price of the item. Moreover, we have a "Stores & Prices" tab, which allows users to view information about their local stores and the item list available at each store. Next, we have a "Routes & Savings" tab, offering three different routes for the user to choose from: The cheapest route, the fastest route, and the most eco-friendly route. Additionally, we offer a "Settings" tab that allows the user to select their shopping preferences and control their account and data. We also provide a chatbot to the user, ensuring streamlined, efficient purchases. When combined, these features turn food sustainability into a personalized, actionable, and rewarding experience.

How we built it

We started by splitting up our team of 4 into two main focuses of 2 people. One group focused on the frontend, using GitHub and React to create a user interface. The other group of 2 people built the backend with FastAPI, where we simulated API data accessing food information and shopping routes. We tied both these parts together to ensure that ShopSmart would be functional and accurate.

Challenges we ran into

While building ShopSmart, the primary issues we encountered arose in the frontend and backend. One of the issues we ran into initially was developing an integrated system that connected our data to our route optimization logic. We found this particularly challenging because our program relied on historical data to provide meaningful, domain-specific, and request-specific diagnoses. Additionally, our search bar also proved to be quite challenging for us to address. While conversing with the AI chatbot or adding new items to our shopping cart, we found that our search bar would often reset itself. This, in turn, caused us to re-evaluate our interface and find another way to add items to the cart. Likewise, due to our short time frame, we were forced to prioritize high-impact features such as rendering map data and calculating cumulative trip costs associated with certain locations and certain products.

Accomplishments that we're proud of

The accomplishments we're proud of are designing a solution that integrates environmental impact with personal utility, ensuring that users are supported in their sustainable shopping. Moreover, we are proud of creating a prototype that demonstrates core functionality with shopping lists, routes, prices, savings, and more, with potential for pragmatic applications in underserved communities. Additionally, we are proud of simulating real data, such as local stores and prices, into actionable, personalized solutions. As a team project, we are proud of the lifelong communication and problem-solving skills we developed in overcoming our aforementioned challenges. From a broader perspective, we are proud of building a strong foundation for a viable product that could scale into a broader sustainability platform for all.

What we learned

What this experience taught us is how to manage effective functionality, code efficiency, and product innovation under pressure constraints induced by time. Specifically, we deepened our understanding of how to use LLMs in real-life cases by connecting a chatbot and training it on an artificial dataset tailored for grocery shopping. Addressing critical front-end issues, such as the glitchy taskbar, we all collectively understood the importance of prioritizing usability and function over aesthetics. By addressing and viewing the discrepancies between modern grocery and inefficient online representation, we highlight our unique end-to-end approach aimed at making this experience easier. Thus, through making a product technically strong and market-ready use-wise, we learned the value of synthesizing different aspects of technology to create optimal results while also ensuring to manage our priorities in terms of efficiency and necessary functionality.

What's next for ShopSmart

In the future and next ventures, we hope to incorporate a feature that allows us to use real-time data, price-wise, in our program. This allows us to adapt to the demands of modern inflation, while keeping data consistent in accordance with the changes that may occur in stores.