Inspiration
Food waste is a significant problem, with many ingredients going unused because people aren't sure what to cook with them. We created PantryPal to help users transform random ingredients into delicious meals through AI-powered recipe suggestions.
What it does
PantryPal is a mobile app that lets users snap photos of their ingredients and instantly receive personalized recipe recommendations. Users simply:
- Take a photo of their ingredients
- Preview and confirm the image
- Get AI-generated recipe suggestions with detailed instructions, cook times, and difficulty levels
How we built it
- React Native with Expo for the mobile frontend
- Express backend for API endpoints and business logic
- Clarifai API for ingredient detection from images
- Cohere for AI-powered recipe generation
- Render for cloud deployment
- TypeScript for type safety
Challenges we ran into
- Poor image detection quality with certain providers (Google Cloud Vision API, Azure Computer Vision)
- Implementing robust error handling for API failures
- CORS issues
Accomplishments that we're proud of
- Built a full-stack AI-powered recipe recommendation system
- Successful integration of multiple AI services (Clarifai, Cohere)
- Clean, responsive mobile UI with intuitive user flow
- Efficient backend architecture handling concurrent requests
What we learned
- Full-stack TypeScript development with React Native and Express
- Integrating and orchestrating multiple AI services (Clarifai, Cohere)
- REST API design and implementation
- Error handling across the stack
- Mobile application deployment
What's next for PantryPal
- User authentication and login
- Functionality for saving and sharing recipes
- Recipe filtering based on preferences
- Training custom models for better food detection
Log in or sign up for Devpost to join the conversation.