Leading AI at Work: Governance, Rollout and Practice / Unit 1 LAI 017 worked example, 60 to 90 seconds
Fitting the Method Into Data, Tests and Prompts
The idea
The same approach extends to data tools, test management and the team's library of AI prompts.
In practice
Step one: add a thing attribute in the data catalog or data models, so anyone can see which tables belong to which thing. Step two: tag each interface with its primary thing, and cite cards in interface documents for the behavior they expect. Step three: tag test cases with their thing and card identifier, so that when a thing changes, the team can query exactly what must be rerun or updated. Step four: keep a small library of prompt templates that name a thing, such as generate test cases for this card using this Pack. Step five: draw the templates from the method's prompt patterns by phase, which The M-CaReM-AI Method, lesson two hundred eight, introduces.
Where people slip
Tags are cheap to add and expensive to retrofit. The book's advice is to tag new work from the day the pilot starts, and to backfill only the high-risk things.
Your place on the trail
Every lesson rests on earlier ones and opens later ones. Tap any to jump.
Unlocks
Something wrong in this lesson? Flag it · Sources for Leading AI at Work: Governance, Rollout and Practice