Inspiration
The inspiration for EcoHero AI stems from a critical environmental crisis: Indonesia ranks as the fifth-largest waste producer in the world and is the second or third-largest contributor of plastic waste to the ocean. Every year, the country generates 24.8 million tons of waste, with roughly 20 million tons entering Indonesian waters However, the core realization driving the project is that the problem is not a lack of education or posters People generally know how to dispose of waste, but they lack consistency, motivation, instant feedback, and a sense of community EcoHero AI was inspired by the need to focus on long-term habit formation rather than just basic education
What it does
EcoHero AI is a "Personal AI Environmental Coach" designed to help users build sustainable habits through vision AI, coaching, and gamification Users simply upload or snap a photo of their waste. In seconds, the AI identifies the waste type (e.g., PET plastic), its recyclability, and its estimated environmental impact (like how long it takes to decompose) More than just an identifier, the "AI Coach" provides personal insights. For example, if it notices 75% of a user's scanned waste is single-use plastic, it will actively recommend bringing a reusable tumbler and calculate the annual positive impact of that change
How we built it
The platform is built on a streamlined technical architecture called the AI Flow: a user's photo passes through a Vision Model for waste detection, which outputs structured JSON data. This data feeds into the AI Coach, updates the user's dashboard, tracks daily challenges, and calculates environmental impact On the front end, the application is built using five strict UX principles, including "One Screen One Purpose" and "AI First" The design draws inspiration from Apple for minimalism, Duolingo for gamification elements (like progress rings and badges), and Headspace for an emotional, calming visual aesthetic
Challenges we ran into
A major design challenge was ensuring Responsible AI. The team had to build a fallback mechanism for when the AI's vision model has a low confidence rate (e.g., below 93%) To solve this, the AI is programmed to honestly admit, "I'm not fully confident. Please check your local recycling guidelines," ensuring the tool is used responsibly Additionally, the team faced the challenge of keeping the user journey incredibly simple so users wouldn't lose interest. They strictly limited the core application flow to exactly six clicks (Landing -> Start -> Scan -> Upload -> Analyze -> Impact -> Leaderboard), entirely removing complex menus, settings, and login screens to reduce friction
Accomplishments that we're proud of
The "Living Earth UI": The team successfully integrated UI storytelling where the dashboard literally comes to life. As users complete challenges, a gray background transforms—greenery appears, virtual trees grow, and birds fly, directly tying digital interactions to environmental healing Creating a Coach, not a Chatbot: The team is proud that the AI delivers highly personalized, actionable cards (like noticing a user's specific plastic habits) rather than relying on a generic conversational chatbot interface
What we learned
The foundational lesson of the project is that information alone does not change behavior. The team learned that to actually solve the waste crisis at an individual level, users require instant visual gratification and a sense of ownership over their environment By instantly updating a mission progress bar the moment a user scans a recyclable item, they learned that immediate feedback is crucial for making users feel the value of their actions
What's next for EcoHero AI: Gamifying the Green Revolution
The ultimate goal for EcoHero AI is to transition from a personal habit-tracking tool into a massive social movement Next steps involve heavily emphasizing community impact and city-based leaderboards The dashboard will show users the power of collective action, displaying messages like: "If 1,000 people complete today's challenge, together we can prevent approximately 5,000 plastic bottles from entering landfills" By pitting cities like Jakarta, Bandung, and Surabaya against each other in eco-challenges, EcoHero AI aims to scale gamification into a nationwide green revolution
Built With
- chatgpt
- codex
- nextjs
- supabase
- typescript
- vercel
Log in or sign up for Devpost to join the conversation.