A tool-library late-notice writer wants to send a new foundation model to every branch because a demo sounded nice. No one has scored it against real overdue letters. What should the team insist on?
Select an answer to reveal the explanation.
Short Explanation
Think of tasting the late-notice letters before you mail them to every branch. Evaluation sits after customization and before production. A polished demo is not a score, and it does not fine-tune the weights.
Full Explanation
In the foundation-model lifecycle, evaluation sits between customization and deployment. A demo that sounded nice is not a score on real overdue letters. Live monitoring cannot replace that pre-production check. A spoken demo does not fine-tune weights.