CCA-F: Claude Certified Architect — Foundations

By Anthropic · Question Mode

QuestionQ3

Agentic Architecture & Orchestration

You’re implementing escalation logic for when the agent should invoke escalate_to_human. Your team has proposed four different ways to trigger escalation. Which approach will most reliably identify cases that truly need human intervention?

Explanation

Human escalation is most reliable when it is triggered by clear, task-relevant conditions: the customer explicitly asks for a human, the required resolution involves a policy exception, or the agent cannot make meaningful progress. These conditions directly reflect a need for human judgment or assistance, unlike sentiment thresholds, fixed retry counts, or rigid category mappings.

Community Discussion

No comments yet. Be the first to start the discussion!