Sparks

Inspiration

Sparks was inspired by our frustration with how difficult it is to find businesses that truly match our personal preferences and values. Whether searching for modest clothing, eco-friendly products, or local artisans, we often found ourselves lost in generic search results dominated by fast fashion and large brands. We wanted to change that by creating a platform that uses AI to connect users with businesses that genuinely reflect what matters to them, making discovery more personal, meaningful, and effortless while giving small businesses the visibility they deserve.

What it does

Sparks is a mobile app that connects shoppers with small businesses aligned with their preferences, like “eco-friendly,” “handmade,” or “BIPOC-owned.” Users can select their preferences, browse recommendations, and spark a connection by swiping right!

How we built it

We learned to use tools like Cloudflare, APIs, and React Native, and to collaborate effectively on GitHub using multiple branches. We also realized the importance of having a clear plan for our tools, structure, and overall approach before starting the building process.

Challenges we ran into

Frontend coding was our biggest challenge. We're all quite new to using React Native, so we struggled a lot in converting our design to code. Frontend was the most time consuming part of Sparks, but we learned a lot to create a seamless user experience.

Accomplishments that we're proud of

Building our first full stack app!

Integration between React Native Expo, Cloudflare, and Gemini AI.

A clean UI that feels professional and easy to use.

What we learned

We learned how to use tools like Cloudflare, AI APIs, and React Native. We also learned how to collaborate better on GitHub with several branches. Our biggest strength was knowing what we were building before we built it, and having the plan for the structure, organization, and tools.

What's next for Sparks

We have some exciting ideas we want to explore, including adding features like business spotlights and user reviews and invest in integrating an MCP for real-time data APIs. We currently have an MVP and hope to build a more seamless, testable UI in the future.