Inspiration

Recycling is something most people know they should do, but knowing what to do and consistently doing it are two different things. We were interested in how technology could make sustainable habits more engaging, especially for younger people.

Rather than creating another platform that simply provides recycling information, we wanted to explore whether AI, gamification, and social accountability could encourage people to take action more consistently. This led to EcoCrew: a social recycling experience where users can complete recycling actions, receive AI-assisted disposal guidance, earn points, and contribute towards shared goals with their crew.

What It Does

EcoCrew turns everyday recycling into a social and rewarding experience.

Users upload a photo of a household item and choose how they believe it should be disposed of. The application provides AI-assisted guidance and feedback to help users make better recycling decisions.

Completing actions contributes towards personal points as well as crew progress. Users can also participate in shared streaks, weekly missions, activity feeds, leaderboards, reactions, and cosmetic rewards.

Our goal was to make sustainable behaviour feel less like an individual chore and more like something people can do together.

How We Built It

We built EcoCrew as a mobile-first web application using Vite, native JavaScript, Bootstrap, and Sass.

For the backend architecture, we integrated Supabase for authentication, user data, profiles, crews, progress, and other persistent application data. We also designed the application to support AI-assisted image analysis through a secure backend architecture using Supabase Edge Functions.

The application was divided into different areas of responsibility so that team members could work in parallel. We developed the core user journey around authentication, onboarding, completing a recycling action, viewing results, and tracking personal and crew progress.

Because reliability is important during a hackathon demonstration, we also created deterministic demo scenarios that allowed the core experience to be demonstrated consistently while backend and AI integrations were being developed.

Challenges We Faced

One of our biggest challenges was managing scope within the limited hackathon timeframe. We started with many ideas around AI verification, social features, gamification, missions, and rewards. The challenge was deciding which features were essential to demonstrating the core idea and which could be simplified.

Another challenge was integrating multiple parts of the application developed in parallel. When different team members work on the frontend, backend, and product features simultaneously, small differences in data structures and expected API responses can quickly cause integration issues. We had to communicate frequently and align our interfaces as the project evolved.

We also faced the challenge of balancing AI capabilities with reliability. AI image analysis can produce inconsistent results, and a hackathon demo needs to remain dependable. This pushed us to think carefully about fallback behaviour and deterministic demo scenarios rather than assuming an AI service would always respond perfectly.

What We Learned

EcoCrew taught us that building a project is not only about writing code. We learned the importance of scoping a problem, prioritising features, designing clear user journeys, and coordinating work across a team.

We also gained a better understanding of how AI features need to be integrated into real applications. AI is most useful when it supports a clear user experience and is combined with sensible validation and fallback behaviour.

Most importantly, we learned that sustainability technology does not always need to solve an entirely new problem. Sometimes, creating impact means finding better ways to motivate people to consistently take actions they already know are important.

EcoCrew is our attempt to make recycling more engaging, social, and rewarding—one action at a time.