EcoRewards

Get rewards for showing your sustainability initiative!

Comment

Inspiration When it comes to climate change, it's easy for individuals to feel like their daily habits don't make a dent. Furthermore, doing the right thing—like meticulously sorting recycling or actively cutting down on electricity—often lacks immediate, tangible benefits. We asked ourselves: What if saving the planet paid off right now? We wanted to bridge the gap between long-term environmental goals and short-term human motivation by gamifying eco-friendly habits and rewarding people for verifiable sustainable actions.

What it does EcoRewards is a gamified sustainability platform that financially and socially incentivizes green habits. Users can earn reward points in two primary ways:

Recycling Verification: Users snap and upload a photo of themselves throwing recyclable items into the correct bins. Our AI verifies the action and awards points.

Utility Bill Reductions: Users upload their monthly utility bills (electricity, gas, or water). If their usage is lower than the previous month, they are rewarded for actively conserving natural resources.

These points can ultimately be redeemed for perks, turning everyday conservation into a rewarding game.

How we built it We engineered a modern, full-stack application designed for speed and scalability:

Frontend: Built with React and TypeScript (TSX) to deliver a clean, responsive, and intuitive user interface.

Backend: Powered by Go (Golang) using the Gin HTTP web framework, ensuring lightning-fast routing and high-performance API handling.

AI Integration: We leveraged the GPT-4o API as our core analysis engine. It handles both the dynamic computer vision required to verify recycling photos (identifying the object and the bin) and the OCR/data-extraction needed to parse and compare utility bills.

Infrastructure: The entire platform and database are hosted on AWS.

Challenges we ran into Time was our biggest adversary. With a smaller team than we ideally needed, connecting all the moving parts—frontend, backend, AI, and cloud infrastructure—under a strict hackathon deadline was intense.

Specifically, configuring our AWS cloud hosting environment took significantly longer than anticipated. We also spent a considerable amount of time testing various models for the photo uploads before realizing GPT-4o offered the exact blend of visual analysis and data parsing we needed to prevent users from easily spoofing the system.

Accomplishments that we're proud of Cloud Deployment under Pressure: Successfully configuring and deploying our database to the cloud despite the severe time constraints.

Dual-purpose AI Accuracy: Fine-tuning our prompts and API calls to make the image recognition incredibly consistent for both messy, dynamic recycling photos and structured, text-heavy utility bills.

What we learned Hackathons are a trial by fire for DevOps! We learned the hard way that cloud configuration almost always takes longer than your initial estimate. On a team management level, we realized the sheer importance of resource allocation; having just one more teammate would have drastically smoothed out the final integration phase.

What's next for EcoRewards Full Integration: Smoothing out the final connections between the frontend and backend and establishing a fully automated CI/CD pipeline on AWS.

Real-world Partnerships: Partnering with local eco-friendly businesses, coffee shops, and transit authorities so users can exchange their in-app points for real-world discounts.

Social Leaderboards: Adding a social component where users can compete with friends, family, or even rival neighborhoods to see who can save the most energy each month.

Expanded Recognition: Upgrading our AI to recognize specialized recycling categories, like composting or e-waste drop-offs.

Built With

Share this project:

Updates