Malaysia is currently facing a significant health challenge, cited as having the highest obesity rates in Southeast Asia. Our culture is built around incredible food, but the hidden caloric impact of local staples like Nasi Lemak and Roti Canai, combined with tropical heat that makes outdoor exercise difficult, leads to rising health risks. I built "Nadi" (Malay for "Pulse") to give Malaysians a culturally-aware AI companion that understands both our diet and our fitness data.
An AI-powered health ecosystem with three core pillars:
- Meal AI: Uses Gemini Flash Vision to identify Malaysian food from photos, providing nutritional breakdowns that recognize local ingredients.
- Run Analyzer: Processes Strava JSON data to provide AI-powered performance insights, fatigue detection, and recovery advice.
- Health Dashboard: Visualizes key metrics like VO2 Max and weekly mileage to track the user's "Nadi" (vitality) over time.
Stack we used:
- Frontend: Next.js
- Backend: Python + FastAPI
- AI Intelligence: Gemini 2.5 Flash
- Database: Supabase for secure storage of health logs and run history.
Log in or sign up for Devpost to join the conversation.