Inspiration
Rising food costs and lack of nutrition knowledge is a global crisis. 30% of food purchased goes to waste, and households struggle to balance health goals with tight budgets. We asked: What if AI could start with your budget, not recipes?
What it does
Cook Healthy AI is an AI-powered meal planning platform that:
- 🧠 Generates personalized weekly meal plans based on your budget first
- 📸 Scans food photos for instant nutrition analysis
- 💬 Provides a conversational AI wellness coach
- 🛒 Creates smart shopping lists with cost estimates
- 📊 Tracks health impact, savings, and cooking streaks
How we built it
- Frontend: Next.js 16, React 19, TypeScript, Tailwind CSS, Framer Motion
- AI: Google Gemini 2.0 Flash for text + vision analysis
- Voice: ElevenLabs for text-to-speech coaching
- Backend: Supabase (PostgreSQL, Auth, Storage)
- Architecture: Budget-first algorithm that maximizes nutrition while minimizing cost
Challenges we ran into
- Gemini Vision integration for accurate food photo analysis
- Building a budget-first AI platform that generates realistic, affordable meal plans
- Internationalization to make the app work globally with USD pricing
- Ensuring graceful fallbacks when AI services are unavailable
What we learned
- Budget constraints should drive meal planning, not the other way around
- Users need visual feedback and gamification to build healthy habits
What's next
- Mobile app with camera integration
- Grocery store API for real-time pricing
- Community recipe sharing and challenges
- Integration with fitness trackers
Built With
- elevenlabs
- framer-motion
- google-gemini-ai
- next.js
- postgresql
- react
- recharts
- supabase
- tailwind-css
- typescript
Log in or sign up for Devpost to join the conversation.