Core | Devpost

Wireframes 1-3
Wireframes 4-7

Inspiration

For many of us, grocery shopping is routine. For immigrant and ESL communities, it can be stressful, confusing, and isolating. Growing up in immigrant families, we’ve seen firsthand how language barriers make everyday tasks like reading labels, comparing products, and identifying allergens unnecessarily difficult, and how that can make people feel excluded from something as basic as food access.

What it does

Core is a multilingual grocery companion that helps users shop with confidence. Users enter their grocery list in their native language, choose their preferences, and Core surfaces relevant products using Kroger data, organized by aisle and category. Each item shows translated names, an image, price, filters like organic or lowest price, aisle location, and allergen information so users only see safe options, with clear warnings for “may contain” items. Users then approve or skip products in-store, making confident decisions as they shop.

How we built it

We built the backend using Python and FastAPI, with PostgreSQL and SQLAlchemy to manage structured product and user data. We integrated Kroger API data to simulate real store inventory and layered translation and filtering logic on top to match user inputs. The frontend was designed in Figma and implemented with React and Tailwind, focusing on clear, accessible UI patterns like category tiles and simple decision flows.

Challenges we ran into

We ran into major challenges on the frontend, especially when translating our Figma designs into working code that aligned with our backend logic. We lost hours trying to bridge the gap between design and functionality, as components that looked simple in Figma proved complex to implement in code. While we had a clear design vision and a functional backend, this integration forced us to iterate quickly, simplify parts of the UI, and make tradeoffs to deliver a working product.

Accomplishments that we're proud of

We’re proud that we built a fully functional end-to-end prototype in such a short time, from user input to a structured, shoppable list. More importantly, we created something grounded in a real problem we’ve seen firsthand, and translated that into a product that feels practical and impactful. The clarity of the user flow and the focus on accessibility made the product feel intentional, not just technical.

What we learned

We learned how to quickly scope and prioritise features to ship something meaningful under tight time constraints. As this was the first hackathon for two of our team members, we were learning as we went and working with new technologies like React, Vite, and Tailwind for the first time. We also gained experience integrating external APIs, structuring backend systems, and designing for clarity and simplicity, especially for users who may already feel overwhelmed.

What's next for Core

Next, we want to expand beyond a limited dataset by pulling in promotional data to highlight deals and savings in real time, while also supporting multiple stores beyond Kroger. We plan to scale accessibility by adding more languages and improving personalisation so the experience better reflects each user’s needs. Longer term, we aim to introduce a conversational chatbot, giving users a comfortable, low-pressure way to ask questions and navigate their shopping experience with confidence.