Inspiration
Food waste is a global issue, yet most of us don’t realize how much we throw away daily. We wanted to build something that brings awareness to personal habits in a fun, interactive way—while promoting sustainability and responsible consumption.
What it does
WasteNot helps users track their daily meals by uploading before-and-after photos. It uses AI to analyze food waste, awards points based on consumption, and gamifies the experience through badges and streaks—making sustainability a part of everyday life.
How we built it
Frontend: React + TypeScript using Vite and Tailwind CSS for a clean and responsive user interface.
Backend: Express server
Image Analysis: Google Gemini and Vision APIs to check for food presence and estimate consumption.
Database: PostgreSQL with Drizzle ORM for meal tracking, user accounts, and rewards.
Authentication: Simple session-based login system (or token-based if implemented).
Challenges we ran into
Fine-tuning Gemini API prompts to consistently analyze food waste accurately.
Ensuring image uploads were validated as actual meals and not just random photos.
Accomplishments that we're proud of
Building a full-stack app with intelligent image analysis in under a weekend
Seamlessly integrating AI to make tracking both smart and user-friendly.
Creating a visually appealing dashboard that encourages sustainable eating habits.
What we learned
How to work with AI APIs for practical use cases.
Gained hands-on experience connecting a React frontend to a Node/Express backend using RESTful routes and managing state with React Query.
Improved our debugging skills by tracing API request flows and handling edge cases in both client and server environments.
What's next for WasteNot
Add nutrition insights and daily goals.
Introduce community challenges and leaderboards.
Expand reward systems with real-world incentives or donations.
Improve AI accuracy with custom training or fine-tuning.
Log in or sign up for Devpost to join the conversation.