Inspiration
Everyone in the world sleeps yet the quality differs for everybody. We wanted to create an app that improves the sleep quality of users.
What it does
This app gives personalized recommendations for the user's day based off of the users previous input day values, such as their hours slept, caffeine intake, activity levels, their day schedule, sleep time.
How we built it
We programmed with AI as a guide to streamline our development process. To begin we found three data sets, 'Student Sleep Patterns', 'Sleep Health and Lifestyle Dataset', 'FitBit Fitness Tracker Data' off Kaggle. We filtered the columns we wanted, and feature engineered new columns that seemed relevant. We trained a Random Forest Machine Learning Model, to predict the user's health score. Using this score, we attached Human Delta to mimic the feature engineering, to do further analysis on the health score and users personalized input values. UI in Swift with SwiftUI—screens, tabs, scrolling stuff, sheets, and the glassy ocean-themed cards are all pretty standard SwiftUI, with shared colors and styles in AppTheme and the oceanCard look. The energy graph uses Apple’s Charts framework. The dashboard’s feedback slider is a tiny bit of UIKit (UISlider through UIViewRepresentable) because SwiftUI’s slider was fighting the scroll view. Combine handles things like timers and refresh loops feeding LocalStore, and Foundation covers models and API plumbing—it’s a straight-up native iOS app, no web UI stack.
Challenges we ran into
Connection between the Human Delta Api key and system code, and UI design challenges. Team constraints, hardware constraints.
Accomplishments that we're proud of
Working with limited time and complete overview of the app.
What we learned
We learned how to use Human Delta, how to narrow down broad concepts, and how to adapt quickly in a short amount of time.
What's next for MyCurrent
More user features such as sound frequencies when you sleep and wearable smart watch integration.
Log in or sign up for Devpost to join the conversation.