Question 100
CCA-F is a newly released exam. These practice questions reflect our best current understanding — the exact wording and question formats may change in the live exam until the certification matures. Use them to learn the concepts, not to memorize.
Scenario: Prompt Engineering and Structured Output - A code-comment checker uses the instruction "Check comments for accuracy and be conservative." It produces many low-value flags (stylistic nitpicks, minor wording) and developers stop trusting it. How should the prompt be rewritten?
- AGive explicit criteria: flag a comment ONLY if it contradicts code behavior, references a non-existent symbol, or a TODO refers to an already-fixed bug—and list what NOT to flag.
- BAdd "be even more conservative and precise" to the instruction.
- CRaise the model temperature to encourage variety in findings.
- DAsk the model to flag everything and let developers filter.