NutriBalance

Inspiration

Non-communicable diseases like diabetes and heart disease are leading causes of death globally, yet many of these conditions are preventable through better nutrition. I wanted to build a tool that makes personalized nutrition guidance accessible to everyone, supporting UN SDG 3's mission for better health and well-being.

What I Learned

Through this project, I deepened my understanding of nutrition science including BMR calculation using the Mifflin-St Jeor equation, TDEE estimation with activity multipliers, and evidence-based macronutrient distribution ranges. I also learned how to structure a Python project with clean architecture separating models, services, and utilities.

How I Built It

I built NutriBalance entirely with Python using a modular architecture. The project has separate layers for data models, business logic, and utilities. All calculations use scientifically validated formulas. I included a comprehensive test suite with 13 unit tests to ensure accuracy.

Challenges

The main challenge was ensuring the nutritional calculations were accurate across different user profiles. Testing edge cases like very low or very high BMI values required careful validation. Setting up the GitHub Actions CI/CD pipeline to run tests automatically was also a learning experience.

Built With

• python