Cura - Smart Supplements

Instant supplements tailored towards your health problem.

Comment

Cura Health App Overview

Inspiration

I were inspired by the overwhelming amount of supplement and wellness information available today. Many people spend hours navigating what to take, in what form, or at what dosage for their specific health needs, then often end up choosing supplements that are mismatched or not optimal for their needs.

With Cura, I aimed to create a reliable, easy-to-use AI assistant that provides informational guidance on supplements, helping users make informed health decisions without replacing professional medical advice.

What it Does

Cura is an AI-powered health app that:

  1. Offers specific informational recommendations on supplements and nutrition.
  2. Explains supplement types, serving sizes, and forms (capsules, liquids, etc.).
  3. Provides a clear, friendly interface for exploring wellness options.

It is designed to be educational and supportive, helping users make informed choices efficiently.

How I Built It

  • Frontend: React with Tailwind CSS for a clean, responsive UI.
  • Animations & Interactivity: Framer Motion for smooth transitions.
  • Icons & Visuals: Lucide React for intuitive, modern icons.
  • Backend: Node.js with MySQL for storing supplement and product data.
  • Deployment: Cloudflare, Render, and Aiven (cloud database).

I use Gemini to analyze user input and suggest relevant supplements based on age, gender, allergies, lifestyle, etc.
Gemini returns a JSON of keywords (e.g., Vitamin A, Zinc), which I then match against our database to provide specific supplement recommendations.

Challenges I Ran Into

  • Data Complexity: Integrating multiple CSV sources and mapping them into the database while ensuring data consistency.
  • AI Limitations: Gemini API has a daily request limit. I sometimes needed more than 4 accounts just to test. TwT
  • Backend Optimization: Millions of supplements in the database required efficient search logic and pre-fetching.
  • UI Design: Simplifying the interface for a smooth user experience while still keeping all the functions and smoothness in the user experience.
  • CSV to Database: Manual import was impossible due to encoding issues, I had to import via code.

Accomplishments

  • Successfully merged multiple complex datasets into a structured, searchable database.
  • Built a clean, interactive, responsive UI making supplement info accessible.
  • Implemented AI-driven guidance while maintaining ethical boundaries.
  • Delivered and deployed a functional MVP demonstrating real-world AI applications in health education.
  • Attending in a hackathon for the first time. This was my first time participating in a hackathon, and it was an incredibly fun experience. I got to see how much can be accomplished with code in a short amount of time and also experienced what it feels like to stay up until 3 am coding. It was truly an honor to attend a hackathon like this, and I am very grateful to the MLH team for organizing such an event.

What I Learned

  • Handling large datasets requires careful planning and validation.
  • Communicating AI limitations is crucial for user trust.
  • Combining backend efficiency with frontend usability ensures a seamless experience.
  • Data modeling and UI design are equally challenging, even for a simple app.
  • Learned how to deploy using Cloudflare and Aiven.

What's Next for Cura - Smart Supplements

  • Implement User System for personalized health tracking.
  • Store search history to provide better, data-driven suggestions.
  • Upgrade Gemini to increase API call limits.
  • Update database weekly based on the active market supplements.

Built With

Share this project:

Updates