NutriMind AI

NutriMind AI was inspired by the growing gap between traditional diet apps and real human behavior. Most platforms focus only on calorie counting, but they ignore emotional eating, stress, and lifestyle habits. We wanted to build something more human-centered that combines nutrition guidance with emotional and mental wellness support.

Through this project, we learned how AI can go beyond simple responses and become a supportive health companion. We also understood the importance of prompt engineering, especially in healthcare, where safety and responsible responses are critical. Designing a system that gives useful but safe advice was a key learning experience for us.

We built the project using React for the frontend, Node.js for the backend, and Gemini API for AI responses. We designed a chatbot-style interface where users enter their mood, habits, and symptoms, and the AI generates personalized diet and wellness suggestions. We also added features like emotional eating detection, smart food swaps, and wellness scoring.

One of the biggest challenges was ensuring the AI responses were both helpful and safe, without giving medical advice. Another challenge was making the system feel personalized instead of generic. We solved this through careful prompt design and iterative testing.

Overall, this project helped us understand how AI can be used in preventive healthcare and how important empathy, simplicity, and responsibility are when designing health-related systems.

Built With

• and
• designed
• for
• healthcare
• nutrition
• prompts
• tested
• we
• we-built-this-project-using-**google-ai-studio-(gemini-api)**-for-the-core-ai-functionality.-the-frontend-was-developed-using-**react.js**-with-**tailwind-css**-for-a-clean-and-responsive-ui-design.-on-the-backend
• we-used-**node.js**-and-**express.js**-to-handle-api-requests-and-connect-the-application-with-the-ai-model.-for-data-handling
• we-used-simple-**json-based-storage-/-local-state-management**-to-keep-the-prototype-lightweight-and-fast.-the-ai-responses-were-powered-through-the-**gemini-api-via-google-ai-studio**
• where