Rating
- size:
3 - complexity:
4
Purpose
Practice identifying hidden review debt and cleanup cost in a team that appears successful if you only look at usage or output volume.
Scenario
A team reports:
- high AI usage
- faster pull request creation
- strong manager sentiment
But there are weaker signals:
- senior reviewers are spending longer in review
- reverted changes increased slightly
- onboarding engineers rarely explain their changes well
- some requirements and test artifacts are more polished but less precise
Task
Use AI to help organize the signals, then decide:
- what looks healthy
- what looks suspicious
- what you would inspect next
Expected output
- hypothesis list about hidden system cost
- recommended sampling plan
- short note on what not to conclude yet
Good AI uses
- organizing qualitative and quantitative signals
- generating inspection questions
- grouping risks by workflow stage
Verification focus
- do not confuse plausible explanations with proof
- keep soft signals separate from validated findings
- identify the smallest next inspection that could reduce uncertainty
Anti-patterns to watch
- defending the rollout because the dashboard is green
- assuming seniors are “just busy”
- treating soft signals as either everything or nothing