← All projects
Client projectApp2023

gRobics

A training app that combines assessment, performance, and live workout data into an individually adapted training plan.

Kotlin MultiplatformSwiftUISensorik

Project and context

gRobics guided users from their initial assessment through active training. Medical and sports-science criteria, personal capabilities, and a performance test formed the basis of an individually adapted training plan.

During training, the app captured performance data and recognized completed exercises and repetitions. Based on this data, it adapted instructions in the active workout and subsequent training sessions automatically.

01

My role

I was responsible for the backend and the shared cross-platform core for iOS and Android. I developed the iOS app together with another frontend developer.

I integrated our Motion AI technology and the Movesense sensor so that heart rate, recognized exercises, and repetitions were available for workout evaluation and plan adjustments.

02

Technical foundations

The central training logic lived in a shared Kotlin Multiplatform module. Assessment, performance, plan generation, and evaluation did not have to be implemented separately for each platform.

The native iOS frontend was implemented in SwiftUI. Ktor handled API communication in the shared core. Bluetooth, the Motion AI SDK, and the Movesense sensor connected heart rate and movement recognition to the active workout.

The training control processed current performance data and automatically adapted the active exercise flow and subsequent training sessions.

03

The central technical challenges

The active workout flow was modeled as a state machine. Exercise changes, breaks, interruptions, and different completion states all had to handle their edge cases cleanly.

At the same time, the flow responded to user input and live sensor data. The training plan, recognized exercise, repetitions, and interface had to remain synchronized.

04

Results and evidence

One shared logic codebase powered both the iOS and Android apps. Both platforms used the same training logic while retaining native interfaces.

The app combined assessment and performance testing with measured performance data. Instructions, training volume, and subsequent sessions could therefore be adapted automatically to the user.

gRobics was released for iOS and Android and worked as intended technically. The product did not establish a lasting position in the market. It was later discontinued and is no longer available in the App Store.

Working on something similar?

Let’s talk about what you need to create.

Let’s talk →