Inspiration
As avid home gardeners frustrated by wasted kitchen scraps and failed compost piles, we saw a gap: composting is full of guesswork for everyday people. We wanted to turn this eco-friendly practice from a chore into an accessible, science-backed process—so CompostHelper was born, built for anyone wanting to reduce waste and grow healthier gardens.
What it does
CompostHelper simplifies home composting with AI-powered item identification (photo-based compostability checks), real-time temperature/humidity monitoring (sensor or manual log), personalized maintenance reminders, step-by-step troubleshooting for common issues, and a directory of local compost drop-off programs. It’s a all-in-one tool for beginners and eco-conscious households alike.
How we built it
We started with user research to map core pain points (unknown compostable items, unmonitored piles, lack of local resources). We built the mobile app with React Native for cross-platform compatibility (iOS 15+/Android 10+), integrated TensorFlow for AI image recognition, used Google Maps API for location services, and Firebase for data storage. Bluetooth Low Energy integration enabled smart sensor connectivity, and Stripe handled premium subscription billing.
Challenges we ran into
Calibrating the AI model for 90% accuracy on common household items took months of testing with diverse waste types. Sensor connectivity inconsistencies across devices required robust error handling. Balancing free/premium features to keep the app accessible while sustaining development was also a tightrope—we landed on a freemium model with meaningful premium perks (unlimited AI checks, sensor integration).
Accomplishments that we're proud of
Hitting a 90% accuracy rate for item identification, building seamless sensor integration for real-time monitoring, and creating a local program directory that covers 95% of urban/suburban areas in the US/EU. The app also maintains a 4.7-star rating, with users reporting 30% less kitchen waste sent to landfills.
What we learned
We deepened our understanding of compost science (thermophilic stage temperature ranges, decomposition triggers) and user-centric design—beginners needed simplified workflows, while experienced composters wanted advanced analytics. We also learned the importance of offline functionality, as many gardeners use the app in yards with spotty internet.
What's next for this project
We’re expanding the AI model to cover more regional waste items (e.g., EU-specific packaging), adding smart home assistant integration (Alexa/Google Home) for voice-based logging, and building a community forum for user-to-user troubleshooting. We also plan to localize the app for non-English European languages and partner with municipal waste programs to expand the local directory coverage.
Log in or sign up for Devpost to join the conversation.