Case study / 2023
Insane AI
Rethinking fitness, systems for engagement.
Team
Design lead & project manager
Brand designer
UX researcher
Background
Insane AI combines social fitness, AI-powered motion tracking, and gamified AR feedback. The product needed to feel energetic and social while keeping workout guidance clear enough to use in motion.
Brand, feedback, and progression needed to work as one movement-first product system.
01 / Brand in UX
Translate energy into usable product language.
Brand attributes became interaction cues, visual hierarchy, motion studies, and feedback states so the app could feel recognisable across workouts, rewards, and social spaces.
02 / Feature groups
Organise complexity around repeat behaviour.
AI feedback, workouts, challenges, routines, progress, and social competition were grouped into insight-led loops: start, guide movement, reward completion, and invite a return.
03 / Style scapes
Align emotion before detailed screens.
Style scapes gave leadership an emotional direction that felt active and ambitious without making every feature high intensity. They also established the visual DNA for a scalable system.
04 / Stakeholder sessions
Make usability decisions visible.
Workshops turned business goals and feature ambitions into flows and wireframes before polish made them expensive to change. That separated workout needs from wider engagement goals.
05 / Prototype loop
Experiment, prototype, test, record, implement.
New motion-led features were treated as hypotheses rather than copied conventions. Stakeholder feedback helped refine ideas before they became reusable system patterns.
06 / Guidelines
Create a product kit built for scale.
Components, assets, and guidelines were organised around clarity, recognition, and repeatable states—giving future workouts and challenges a foundation instead of a fresh visual problem.
07 / System in motion
Turn guidelines into product assets.
Static rules, motion studies, and UI components needed to connect so new workout and challenge surfaces could be built with the same visual and behavioural language.
08 / Interaction framework
Guide before, during, and after movement.
The interaction framework defined what the UI should do around a workout: prepare the user, make feedback fast to parse during motion, and reward the result afterwards.
09 / Progression
Carry the brand through progress mechanics.
Levels, achievement language, and analytics gave people an understandable sense of where they were and why another session mattered—without reducing progress to a status label.
10 / Social space
Keep the workout central.
Social mechanics were shaped around participation, comparison, and return motivation rather than feed consumption. Community features had to support movement rather than distract from it.
11 / Routine levels
Unlock difficulty without overwhelm.
The level system exposed more demanding routines gradually, allowing beginners to start clearly and advanced users to follow visible progression paths with practical purpose.
12 / Product direction
A coherent foundation for fitness feedback.
The final direction united brand, real-time feedback, routines, social motivation, and product growth into one scalable system rather than a collection of isolated screens.
