Inspiration
Food waste and improper waste disposal are everyday problems that affect both the environment and communities. We wanted to build one intelligent platform that helps people understand where their waste belongs while also helping them reduce food waste before it happens. This inspired us to create EcoSort AI.
What it does
EcoSort AI is an AI-powered waste management and food sustainability platform. Users can upload a photo of waste and receive instant classification into recycling, compost, landfill, e-waste, or hazardous waste, along with disposal guidance and sustainability tips.
It also includes a digital fridge, expiry tracking, AI recipe recommendations, meal planning, grocery lists, surplus food sharing, social features, and restaurant discovery.
How we built it
We built EcoSort AI using Python, FastAPI, Uvicorn, Vanilla JavaScript, CSS, SQLite/MongoDB, OpenAI, Google Gemini, and HuggingFace.
AI vision models analyze uploaded images and classify waste or identify food ingredients. HuggingFace models power recipe generation and embeddings, while our backend connects the AI features with the web interface and data storage.
Challenges we ran into
One of our biggest challenges was making image-based classification reliable across different lighting conditions, objects, and image qualities. We also had to design fallback systems for AI models and recipe generation so the platform could remain useful when a particular model or API was unavailable.
Integrating multiple AI services into one simple user experience was another major challenge.
Accomplishments that we're proud of
We are proud to have built a complete sustainability platform rather than a single-purpose classifier. EcoSort AI combines waste classification, food management, recipe generation, meal planning, and surplus sharing in one ecosystem.
We are especially proud of turning an uploaded image into practical, actionable sustainability guidance within seconds.
What we learned
We learned how to integrate multiple AI models and APIs into a single application, build reliable fallbacks, design AI-powered user experiences, and connect environmental sustainability with practical everyday tools.
Most importantly, we learned that AI becomes more useful when it doesn't just identify a problem—it helps users take the next step.
What's next for ECOSORT-AI
Our next goal is to make EcoSort AI more intelligent, localized, and community-driven. We plan to improve waste classification accuracy, add multilingual support, expand the surplus food network, introduce smarter expiry predictions, and develop partnerships with restaurants, grocery stores, recycling centers, and local communities.
Ultimately, we want EcoSort AI to become a complete digital platform for reducing waste, saving food, and making sustainable choices easier for everyone.
Log in or sign up for Devpost to join the conversation.