Engineering leads or managers need to understand how their team uses AI to write code and assess that usage when reporting upward or making tooling decisions.
The current approach is to judge a team's AI usage roughly from code review, commit history or verbal reports.
The candidate material does not say what specific pain assessing AI usage causes today, nor what is lost by not assessing it.