Inspiration
We've all bought a bottled drink before - whether that's a water bottle, soda, or juice - but few of us realize the compound environmental cost behind that purchase. Americans alone consume over 50 billion plastic bottles annually, with less than 30% being recycled. Beyond taking hundreds of years to decompose, these bottles generate concerning levels of CO2 emissions before you even open the bottle.
The lack of transparency in the bottled drink supply chain inspired me to create Trace, an app that traces exactly where your plastic bottle came from and calculates the carbon footprint emitted throughout its entire journey.
What it does
Trace uses real-world supply chain data to map the complete journey of your plastic bottled water from source to consumer. Users simply input their location and select a water brand, and Trace will show an interactive map showing the path from municipal water source --> treatment facility --> bottling plant --> distribution center --> your location along with CO2 emissions for each segment. For any missing location data (e.g. municipal water sources), Trace uses Gemini to as an agent to search the web and your map for the right locations.
How we built it
Frontend: Next.js + TypeScript + Tailwind CSS. Mapbox for interactive mapping Backend: Express.js server with a MongoDB database and Auth0 for authentication AI Integration:Google's Gemini 1.5-flash
Challenges we ran into
Programming the agent to accurately get real and correct locations was more difficult than I anticipated. I had to implement multiple steps, including searching the web and verifying with Google Maps and credible sources.
Accomplishments that we're proud of
Soloing this project in < 36 hours!
What I learned
How to use AI for geospatial intelligence
What's next for Trace
- Tell you which water bottle brand would have the least carbon footprint before you buy it
- Go beyond water to include sodas, juices, and other drinks with their unique supply chains and ingredients

Log in or sign up for Devpost to join the conversation.