P1 · 8 questions

Solution Design & Architecture — Attempt 2

CCAR-P P1 practice. Some items are “select all that apply” — they tell you how many to pick and are graded all-or-nothing. Commit before revealing, then read every explanation.

0 of 8 answered

P1 · Q1The agentic loop and when autonomy helps
When is granting an agent an autonomous plan-act-observe loop most warranted?
P1 · Q2Prompt-chaining vs routing vs parallelization
Incoming support tickets must each be sent to one of three specialized handlers based on the ticket's category. Which workflow pattern fits?
P1 · Q3Adapting Claude without fine-tuning
A customer wants Claude to follow their domain terminology and answer from their private knowledge base. They ask how to fine-tune the model on their data. What should you advise?
P1 · Q4Sync vs async/batch, SLA-driven
A nightly job re-scores ten million archived documents for relevance. There is no user waiting; results are consumed the next morning, and cost must be minimized. Which execution mode is most appropriate?
P1 · Q5Tool granularity and the deterministic-vs-model boundary
An agent computes sales tax by reasoning about rates in natural language, and the totals are occasionally wrong. Regulators require the calculation to be exact every time. What is the best next action?
P1 · Q6Context isolation and context bleed
A coordinator delegates two unrelated subtasks to two subagents. The first subagent's private scratch notes are appearing in the second subagent's output. What is the root cause?
P1 · Q7Orchestration topologiesSelect 2
Which two pairings of orchestration topology to use-case are correct? (Select 2)
P1 · Q8Capacity, rate limits, and token budgetingSelect 2
A launch expects a large traffic spike that may exceed the account's rate limits. Which two capacity design choices are sound? (Select 2)

More P1 attempts