FITMIND AI

Most fitness trackers focus purely on physical metrics like steps or calories. I was inspired to build FitMind AI because I realized that our emotional state is the biggest factor in workout consistency. If you’re feeling "alone" or "tired," a high-intensity interval training (HIIT) session might feel impossible, but a walk or a stretch could be exactly what you need to stay active. How I Built ItI developed this prototype using SwiftUI for a native iOS experience.Logic Engine: I created a custom analyzeFeeling() function that parses user input to map emotions to specific health recommendations.Data Visualization: I integrated the Swift Charts framework to transform raw mood scores into a visual history.Architecture: The app uses a structured MoodEntry model to ensure data consistency across the session. What I LearnedI gained deep insights into State Management in SwiftUI, specifically how to use @State to create a reactive UI that updates instantly when the "AI" provides a suggestion. I also learned how to implement BarMark charts to provide users with immediate, actionable feedback on their emotional trends. Challenges FacedThe primary challenge was creating a logic flow that felt "smart" despite being a prototype. I had to balance simple keyword matching with a scoring system to quantify abstract feelings. For example, I assigned numerical values to moods to calculate the "Mood Score" $S$ for the graph:$$S = \sum_{i=1}^{n} \text{mood_weight}_i$$Ensuring the ScrollView handled the dynamic growth of the chart without breaking the layout required careful use of VStack spacing and frame constraints.

Built With

• api
• charts
• data
• ios
• modelling
• swift
• swiftui
• xcode