The swipe gesture’s simplicity and efficiency made it an ideal choice for training the AI. Drawing from Tinder’s success in gamifying decision-making, we recognized that this interaction could be repurposed to create a seamless and engaging way for users to provide feedback to the AI. By mirroring real-world actions like sliding cards, we ensured a natural and familiar experience.

Early testing revealed that users were more likely to interact with the system using swipe gestures compared to traditional buttons. Feedback highlighted the swipe’s simplicity and its contribution to maintaining a flow state during repetitive tasks.
The click prototype was tested at The Company’s workshop locations in Berlin, Hamburg, Cologne, and Munich. During these sessions:
Maintenance staff provided feedback on usability, navigation, and workflow efficiency. I attended the testing in Berlin to observe interactions and document improvements.

By adopting the swipe gesture, we transformed what could have been a mundane task into an engaging and purposeful activity. This approach not only improved the user experience but also served as a critical foundation for the continuous training and improvement of our AI model.

The project began with an in-depth review of a requirements document comprising over 2,000 points. Key insights included:
User Roles: Identifying various stakeholders, including maintenance workers, team leads, and workshop managers. Pain Points: Complex workflows, communication gaps, and lack of real-time fault tracking were critical challenges.

Based on the requirements analysis, I developed low-fidelity wireframes that:
Simplified complex processes into intuitive workflows. Incorporated core functionalities such as fault logging, categorization, and status tracking. Ensured compatibility with desktop and mobile devices used across workshops.
These wireframes were iteratively refined through feedback from The Company’s internal project team.