Enabling self-inspection of vehicles requires a remote inspection tool to validate that imagery. We saw this as an opportunity to also experiment with AI damage detection.

I designed a remote inspection tool that enabled human inspectors to validate and cost the damage detected in photos from customer self-appraisals.

Where applicable inspectors could also validate damage detected by AI and train it on how to correctly cost repairs. The tool was designed to be supplier agnostic and functioned with or without AI-assistance.

It was widely used by inspectors and provided the business with a platform to test the efficacy of a number of AI suppliers.