Inspiration
I wanted to create something useful while learning AI integration, APIs, and deployment. FitAI started as a personal project to manage my own fitness, nutrition, and mental wellness, and over time evolved into a general tool anyone can use.
What it does
FitAI generates personalized health plans for users, including:
- Meal plans with instructions
- Workout plans with instructions
- Skin, hair, and personal care suggestions
- Music generation using the Sonauto API
- Lifestyle and medication advice
- A to-do tracker to help manage daily routines The app is designed to help users maintain mental and physical health in a structured, AI-assisted way.
How i built it
I began FitAI with a rough design that evolved as I built it. First, I worked on the frontend using React/Next.js to create a clean and simple interface. Once the UI was ready, I developed the backend with Node.js and Prisma, then connected a PostgreSQL database hosted and deployed on Neon.
After setting up the core structure, I integrated Google AI Studio and Sonauto API using POST API calls to generate personalized meals, workouts, and music. The project was deployed on Vercel by linking my GitHub repository and Neon database.
I tested everything step by step, debugging API and database connection issues with help from Cursor and ChatGPT. The best moment was when the AI APIs and database connected successfully — that’s when FitAI truly came to life.
Challenges i ran into
Connecting the PostgreSQL database hosted on Neon to the app deployed on Vercel required careful configuration and debugging. Another minor issue is that fetching results from AI APIs currently takes around 1.5 minutes due to free-tier limitations.
What I learned
- How to integrate AI APIs into a real application
- Using OAuth for secure authentication
- Deploying a PostgreSQL database on Neon and connecting it to a deployed app on Vercel
What's next for FitAi
- Reduce API response time by optimizing queries or upgrading API plans
- Add food images and GIF videos for exercises
- Implement ordering for meals, equipment, and medications
Built With
- google-ai-studio
- javascript
- postgresql
- prisma
- react/next.js
- sonauto-api
- typescript
