Inspiration

Tracking nutrition in real life is difficult. Most people do not measure ingredients, calculate calories, or maintain food logs consistently because the process is slow and repetitive. Existing apps often require manual entry and become hard to sustain.

We wanted to build a smarter and easier solution where users can simply upload a food image, enter a recipe, or use grocery data to receive useful nutrition insights instantly.

That idea led to NutriBuddy — an AI-powered nutrition assistant focused on convenience, personalisation, and healthier daily decisions.


What it does

NutriBuddy helps users understand and improve their eating habits using AI.

Main features include:

  • Food image analysis for estimated calories and nutrients
  • Recipe analysis and Healthify recipe suggestions
  • Multiple-frame recipe analysis for better accuracy.
  • AI nutrition chat assistant
  • Family nutrition profiles under one app
  • Smart grocery list and grocery cart insights
  • Supplement optimizer based on dietary gaps
  • Goal-based personalized recommendations

NutriBuddy is designed for individuals as well as families.


How we built it

We built NutriBuddy using Flutter for a cross-platform mobile experience.

Core technologies used:

  • Flutter & Dart for frontend development
  • Firebase Authentication for secure login
  • Cloud Firestore for storing user profiles and goals
  • Google Gemini API for image and text intelligence
  • AMD-focused hybrid inference concept for local accelerated AI processing with cloud fallback
  • State management and reusable Flutter widgets for scalable UI

The system takes user inputs such as food photos, recipes, goals, and shopping preferences, then converts them into personalised health recommendations.


Challenges we ran into

  • Estimating nutrition from mixed food dishes
  • Handling diverse cuisines and regional meals
  • Balancing speed, cost, and AI accuracy
  • Designing a clean UI with many features
  • Managing API limits during testing
  • Creating meaningful family and grocery workflows

These challenges helped us improve the architecture and user experience.


Accomplishments that we're proud of

  • Built a complete AI-powered nutrition ecosystem instead of a single-feature app
  • Combined image analysis, recipe intelligence, shopping, family wellness, and personalisation
  • Designed a scalable mobile app using Flutter + Firebase
  • Integrated modern AI capabilities into a real-world healthcare problem
  • Created a strong prototype with practical use cases

What we learned

Through this project, we learned:

  • How to build real-world AI applications
  • Flutter app architecture and state handling
  • Firebase authentication and database integration
  • Prompt engineering for nutrition use cases
  • Product thinking and user-first feature design
  • How to turn hackathon ideas into working prototypes

What's next for NutriBuddy

Our future roadmap includes:

  • Wearable device integration
  • Deeper Indian food dataset support
  • Barcode and receipt scanning
  • Real-time grocery app extensions
  • More advanced local AMD GPU inference
  • Doctor/dietician collaboration tools
  • Community challenges and gamification

NutriBuddy aims to make healthy living easier, smarter, and more accessible for everyone.

Built With

Share this project:

Updates