Food Waste Scanner

AI-powered recipe scanner that detects food waste risk and offers smart storage tips.

Comment

Inspiration

Food waste is a massive yet often overlooked problem. We were inspired by the everyday frustration of tossing out spoiled ingredients — herbs that wilt too soon, forgotten veggies, or dairy that goes bad before it's used. We aimed to create a straightforward tool that enables individuals to be more mindful in the kitchen and reduce unnecessary waste by leveraging the power of AI.

What it does

Food Waste Scanner is an AI-powered web app that analyzes any recipe and identifies which ingredients are likely to be wasted. It highlights high-risk items and offers personalized suggestions to store, preserve, or repurpose them. Users can paste, speak, or upload their recipe — and even chat with a smart assistant for follow-up questions, such as “How do I store parsley?” or “What can I make with extra mushrooms?”

How we built it

We used:

  • Flask (Python) for the backend
  • OpenAI API to analyze ingredients and suggest waste-reduction tips
  • HTML, CSS, JavaScript, and Bootstrap for a responsive frontend
  • PDF.js for parsing uploaded recipe PDFs
  • Web Speech API for voice input
  • JavaScript & Flask integration for dynamic chat and user feedback
  • All input is analyzed asynchronously with clear feedback, and AI responses are handled via structured JSON.

Challenges we ran into

  • Parsing unstructured recipe text reliably (especially from PDFs and speech)
  • Migrating from older OpenAI completion models to the new openai.ChatCompletion format
  • Ensuring a clean and intuitive UI while integrating multiple inputs (text, file, voice)
  • Managing state for user input and chat history in a single-page flow

Accomplishments that we're proud of

  • Built a full-stack AI app from scratch in under 3 weeks
  • Created a clean, accessible UI that works across devices
  • Enabled voice input, real-time feedback, and interactive chat
  • Seamlessly integrated AI suggestions with a real use case: reducing food waste

What we learned

  • Prompt engineering and JSON-based AI responses
  • File upload handling and live PDF preview in Flask
  • UI/UX best practices for showing AI status, results, and errors
  • The power of small, helpful AI tools in driving behavior change

What's next for Food Waste Scanner

  • Recipe suggestions based on leftover or high-risk ingredients
  • User history & profiles to track saved recipes and suggestions
  • Multi-language support for wider accessibility
  • Mobile app version with offline capabilities
Share this project:

Updates