Anthropic CCAR-P (Claude Certified Architect - Professional) Practice Test 53

Question 520

A backend team asks Claude to implement a billing calculation with several edge cases and regulatory rounding rules. Which choice most directly satisfies the documented goal?

  1. Use test-first acceptance cases for generated changes
  2. Run reproducible CI checks on AI-assisted code changes
  3. Package reusable engineering instructions as governed Skills
  4. Break large refactors into independently reviewable changes

Explanation

Correct Answer

A:

Define executable tests and expected behavior before relying on AI-generated implementation, then review failures and iterate; test-first evidence anchors generated code in independently defined behavior; evaluate the design with passing unit and property tests including specified rounding boundaries.

Incorrect Answers

B:

Execute standardized linting, type checks, tests, and security scanning in versioned CI environments before merges; local success can be environment-dependent and should not be the only release assurance; this alternative would measure cI pass reproducibility on clean builds with locked dependencies; it does not establish passing unit and property tests including specified rounding boundaries for the described use case.

C:

Maintain task-scoped reusable Skills or playbooks with versioning, ownership, and documented prerequisites; shared instructions improve consistency but do not replace enforced approval or automated tests; this alternative would measure checklist completeness across independent sessions using the reviewed Skill version; it does not establish passing unit and property tests including specified rounding boundaries for the described use case.

D:

Use small scoped changes, compatibility checkpoints, and separate pull requests tied to specific tests; one massive patch can reduce initial coordination but increases regression and rollback risk; this alternative would measure review time and defect localization per incremental change set; it does not establish passing unit and property tests including specified rounding boundaries for the described use case.

Question 521

Requiring every description to be manually approved removes the productivity benefit despite low demonstrated risk. How should the design be corrected?

  1. Use risk-based sampling for low-impact decisions
  2. Require named authorization before moving money
  3. Give reviewers only necessary sensitive information
  4. Require stronger approval paths for exceptional high-impact actions

Explanation

Correct Answer

A:

Combine automated checks with representative human sampling and exception-based review scaled to risk; sampling effectiveness must balance residual errors and human capacity; verify the repair using defect-detection rate and reviewer workload under a risk-tiered sampling policy.

Incorrect Answers

B:

General user authentication does not confer approval for a specific financial transfer; separate recommendation from execution and require an authenticated human approval bound to the exact amount, payee, and transaction; that control is assessed by zero disbursements without an auditable transaction-specific approver; the incident instead requires defect-detection rate and reviewer workload under a risk-tiered sampling policy.

C:

Human oversight is not permission to expand access to unrelated personal information; minimize, redact, and scope information in review interfaces according to purpose and authorized reviewer role; that control is assessed by unauthorized field exposure and successful review accuracy after data minimization; the incident instead requires defect-detection rate and reviewer workload under a risk-tiered sampling policy.

D:

A uniform low-friction approval does not reflect differing blast radius and operational risk; apply risk-tiered approvals, including dual control or independent sign-off where policy demands, to exceptionally consequential mutations; that control is assessed by critical-action dual-approval compliance and zero bypasses during staged tests; the incident instead requires defect-detection rate and reviewer workload under a risk-tiered sampling policy.

Question 522

Finance is comparing two configurations for a document assistant: one has cheaper API calls but more manual cleanup. Which denominator and cost components produce the most defensible economic comparison?

  1. End-to-end completed cases per paid work-hour with rework and appeal volumes
  2. Successful completion of critical workflows during injected model outages
  3. Total operational cost divided by approved documents meeting the quality rubric
  4. Incremental resolution lift compared with a comparable non-assisted cohort

Explanation

Correct Answer

C:

Unit cost tied to accepted results distinguishes economical quality from cheap failed attempts; measure total operational cost divided by approved documents meeting the quality rubric at the release gate; combine inference, retrieval, tools, review, and rework costs into cost per accepted deliverable.

Incorrect Answers

A:

Net throughput reflects business output rather than the volume of AI text produced; measure net time saved after review, appeals, and escalations, segmented by complexity, before claiming productivity; this tests end-to-end completed cases per paid work-hour with rework and appeal volumes; the actual release condition is total operational cost divided by approved documents meeting the quality rubric.

B:

Fallback drills demonstrate that the business process is not hostage to one generative dependency; separate critical deterministic operations from AI enhancement and provide explicit fallback and manual paths when model service is unavailable; this tests successful completion of critical workflows during injected model outages; the actual release condition is total operational cost divided by approved documents meeting the quality rubric.

D:

A defensible counterfactual is required to credit the architecture with the observed benefit; use matched baselines, controlled rollout, or appropriate difference-in-differences analysis to isolate contribution under plausible confounders; this tests incremental resolution lift compared with a comparable non-assisted cohort; the actual release condition is total operational cost divided by approved documents meeting the quality rubric.

Question 523

Synchronous processing creates burst throttling and expensive retries during office hours. Which action best removes the underlying failure mode?

  1. Parallelize independent retrieval calls on the critical path
  2. Direct straightforward requests to a validated economical model
  3. Schedule non-urgent work through bounded batch processing
  4. Reuse expensive stable prefixes through prompt caching

Explanation

Correct Answer

C:

Use suitable batch orchestration and queue control to trade immediate response time for predictable throughput and expense; deadline-oriented throughput is the relevant performance objective for noninteractive work; verify the repair using cost and successful throughput per nightly processing window.

Incorrect Answers

A:

Parallelism adds little if operations depend on each other or contend for the same rate-limited service; parallel retrieval with per-call budgets reduces wall time while preserving authorization and complete aggregation; that control is assessed by p95 response time with all three source checks present under load; the incident instead requires cost and successful throughput per nightly processing window.

B:

Using the cheapest model universally could impair complex or safety-critical cases; evaluate a smaller model on straightforward request strata and escalate only where measurable capability is needed; that control is assessed by accepted resolution rate and cost per case across routed difficulty levels; the incident instead requires cost and successful throughput per nightly processing window.

D:

Selecting a smaller model alone may reduce price but does not address wasted recomputation of invariant content; keep a stable eligible prompt prefix and measure cache-read tokens to reduce repeated processing; that control is assessed by cache-read rate, p95 latency, and cost per accepted answer under repeated queries; the incident instead requires cost and successful throughput per nightly processing window.

Question 524

Input processing costs remain high because identical boilerplate is sent without any effective cache reuse. Which action best removes the underlying failure mode?

  1. Process non-urgent jobs using bounded offline batches
  2. Parallelize independent retrieval calls on the critical path
  3. Reuse expensive stable prefixes through prompt caching
  4. Reduce context using retrieval and relevance filtering

Explanation

Correct Answer

C:

Keep a stable eligible prompt prefix and measure cache-read tokens to reduce repeated processing; joint economics and quality checks demonstrate that reuse helps without sacrificing correctness; verify the repair using cache-read rate, p95 latency, and cost per accepted answer under repeated queries.

Incorrect Answers

A:

Batch pipelines are inappropriate where a person needs a subsecond answer for an active decision; use suitable batch orchestration and queue control to trade immediate response time for predictable throughput and expense; that control is assessed by cost and successful throughput per nightly processing window; the incident instead requires cache-read rate, p95 latency, and cost per accepted answer under repeated queries.

B:

Parallelism adds little if operations depend on each other or contend for the same rate-limited service; parallel retrieval with per-call budgets reduces wall time while preserving authorization and complete aggregation; that control is assessed by p95 response time with all three source checks present under load; the incident instead requires cache-read rate, p95 latency, and cost per accepted answer under repeated queries.

D:

Token truncation at arbitrary character positions can remove governing exceptions and worsen accuracy; retrieve relevant excerpts and retain enough surrounding evidence for interpretation rather than indiscriminately filling the context window; that control is assessed by supported-answer rate per input token on labeled audit questions; the incident instead requires cache-read rate, p95 latency, and cost per accepted answer under repeated queries.

Question 525

An employee-policy assistant replaces full-manual prompts with targeted evidence retrieval. Evaluators need to show that context shrinks without losing support for answers. Which measurement best demonstrates the trade-off?

  1. Measured current-state citation accuracy across log-update sequences
  2. Citation correctness by effective-date version in temporal regression questions
  3. Answer support and retrieved-token volume on a sampled policy-question set
  4. Correct recovery of approvals and pending actions after forced session reset

Explanation

Correct Answer

C:

Relevance per token measures whether context compression preserves the needed evidence; measure answer support and retrieved-token volume on a sampled policy-question set at the release gate; apply query-driven retrieval and evidence selection before prompting Claude, retaining citations and permission filters.

Incorrect Answers

A:

Freshness-aware tests expose whether stale tool observations override newer facts; prune or summarize superseded tool data, preserve timestamps and current authoritative observations, and fetch details only when needed; this tests current-state citation accuracy across log-update sequences; the actual release condition is answer support and retrieved-token volume on a sampled policy-question set.

B:

Metadata-aware checks distinguish legitimate historic references from current-policy answers; supply source identity, version, effective date, and retrieval relevance alongside each chunk so the model can attribute evidence correctly; this tests citation correctness by effective-date version in temporal regression questions; the actual release condition is answer support and retrieved-token volume on a sampled policy-question set.

D:

Durable-state testing proves behavior survives context eviction; store approvals and workflow state in an authoritative durable system; treat chat history as context, not a transaction ledger; this tests correct recovery of approvals and pending actions after forced session reset; the actual release condition is answer support and retrieved-token volume on a sampled policy-question set.

Question 526

A caseworker copilot drafts rapidly but increases reviewer corrections for difficult cases. Which design decision directly addresses this constraint?

  1. Translate response-time expectations into a percentile SLO
  2. Use a control group to establish incremental business impact
  3. Design a degraded mode that protects the value proposition
  4. Measure productivity after correction and escalation

Explanation

Correct Answer

D:

Measure net time saved after review, appeals, and escalations, segmented by complexity, before claiming productivity; net throughput reflects business output rather than the volume of AI text produced; evaluate the design with end-to-end completed cases per paid work-hour with rework and appeal volumes.

Incorrect Answers

A:

Define a p95 or p99 target against real end-to-end requests, including retrieval and tools, rather than reporting an average model time; a percentile SLO does not alone reveal whether the answers are factually correct; this alternative would measure end-to-end p95 latency and error rate against the agreed service objective; it does not establish end-to-end completed cases per paid work-hour with rework and appeal volumes for the described use case.

B:

Use matched baselines, controlled rollout, or appropriate difference-in-differences analysis to isolate contribution under plausible confounders; before-after counts without a credible comparison group may misattribute unrelated changes to the model; this alternative would measure incremental resolution lift compared with a comparable non-assisted cohort; it does not establish end-to-end completed cases per paid work-hour with rework and appeal volumes for the described use case.

C:

Separate critical deterministic operations from AI enhancement and provide explicit fallback and manual paths when model service is unavailable; high answer accuracy in normal mode says little about process continuity during dependency failures; this alternative would measure successful completion of critical workflows during injected model outages; it does not establish end-to-end completed cases per paid work-hour with rework and appeal volumes for the described use case.

Question 527

Changing an ID in a request retrieves an invoice belonging to another customer organization. Which intervention targets the defect rather than its symptoms?

  1. Verify tenant ownership of resource identifiers
  2. Issue short-lived scoped credentials for tools
  3. Require separate approval for high-impact mutations
  4. Record the user, tenant, and action in audit trails

Explanation

Correct Answer

A:

Check the requested resource against the authenticated tenant and role at the data service rather than trusting user-provided identifiers; object-level access checks close the horizontal privilege gap that prompt filtering misses; verify the repair using denial of cross-tenant identifier enumeration attempts in direct tool tests.

Incorrect Answers

B:

Encrypting a shared permanent key at rest does not mitigate broad privilege after the key is exposed; prefer tightly scoped, expiring credentials with secure rotation and audited issuance for the precise operation; that control is assessed by credential-expiry and scope-denial tests after simulated token compromise; the incident instead requires denial of cross-tenant identifier enumeration attempts in direct tool tests.

C:

A general sign-in session does not imply user consent to every destructive operation in that session; bind high-impact actions to a separately verified approval and narrowly scoped execution credential; that control is assessed by zero critical network changes without a recorded action-specific approval; the incident instead requires denial of cross-tenant identifier enumeration attempts in direct tool tests.

D:

Logging only a model response cannot attribute which authenticated principal caused a backend change; propagate authenticated actor, tenant, delegated scope, request ID, and action outcome into tamper-resistant audit records; that control is assessed by audit reconstruction of actor, authority, target, and result for sampled mutations; the incident instead requires denial of cross-tenant identifier enumeration attempts in direct tool tests.

Question 528

A risk assistant includes entire audit reports even when the request concerns one control. Which intervention is most appropriate for this use case?

  1. Schedule non-urgent work through bounded batch processing
  2. Route straightforward requests to a validated economical model
  3. Reduce context using retrieval and relevance filtering
  4. Reuse expensive stable prefixes through prompt caching

Explanation

Correct Answer

C:

Retrieve relevant excerpts and retain enough surrounding evidence for interpretation rather than indiscriminately filling the context window; evidence efficiency tests whether trimming preserves the information that matters; evaluate the design with supported-answer rate per input token on labeled audit questions.

Incorrect Answers

A:

Use suitable batch orchestration and queue control to trade immediate response time for predictable throughput and expense; batch pipelines are inappropriate where a person needs a subsecond answer for an active decision; this alternative would measure cost and successful throughput per nightly processing window; it does not establish supported-answer rate per input token on labeled audit questions for the described use case.

B:

Evaluate a smaller model on straightforward request strata and escalate only where measurable capability is needed; using the cheapest model universally could impair complex or safety-critical cases; this alternative would measure accepted resolution rate and cost per case across routed difficulty levels; it does not establish supported-answer rate per input token on labeled audit questions for the described use case.

D:

Keep a stable eligible prompt prefix and measure cache-read tokens to reduce repeated processing; selecting a smaller model alone may reduce price but does not address wasted recomputation of invariant content; this alternative would measure cache-read rate, p95 latency, and cost per accepted answer under repeated queries; it does not establish supported-answer rate per input token on labeled audit questions for the described use case.

Question 529

A procurement steering group wants monthly summaries of business gains, unit costs, and safety performance from a deployed copilot. Which result would show that the agreed indicators are delivered in a form suitable for decisions?

  1. Time to appropriate escalation and fallback activation during drills
  2. Time from actionable feedback to triage, owner assignment, and verified fix
  3. Affected-user awareness and support-ticket volume following staged changes
  4. Delivery of agreed business and risk indicators on schedule with decision-ready interpretation

Explanation

Correct Answer

D:

Reporting completeness lets stakeholders govern the system using relevant evidence; measure delivery of agreed business and risk indicators on schedule with decision-ready interpretation at the release gate; agree on outcome-aligned service reports, cadence, escalation thresholds, and audiences during solution definition.

Incorrect Answers

A:

Incident response tests verify that operational ownership works under pressure; define incident severity, on-call ownership, manual fallback authority, and customer communication triggers; this tests time to appropriate escalation and fallback activation during drills; the actual release condition is delivery of agreed business and risk indicators on schedule with decision-ready interpretation.

B:

Issue lifecycle metrics show whether feedback produces controlled improvement; classify feedback by severity and root-cause type, assign owners, link trace IDs, and feed validated cases into regression tests; this tests time from actionable feedback to triage, owner assignment, and verified fix; the actual release condition is delivery of agreed business and risk indicators on schedule with decision-ready interpretation.

C:

Communication quality is reflected in predictable adoption and fewer avoidable surprises; define release notes, behavior-change notices, affected audiences, and support channels proportional to the deployment's impact; this tests affected-user awareness and support-ticket volume following staged changes; the actual release condition is delivery of agreed business and risk indicators on schedule with decision-ready interpretation.

Question 530

A payment-credit tool times out after committing a write. The caller retries without an operation key and double-posts the credit. Which state-management mechanism prevents this outcome?

  1. Capture reviewer corrections as labeled feedback
  2. Persist workflow state and make retries idempotent
  3. Link each output to source and processing trace IDs
  4. Use a queue and asynchronous workers for bursty workloads

Explanation

Correct Answer

B:

Persist state transitions and use idempotency keys at side-effect boundaries to avoid duplicate transactions across replays; replay tests directly exercise the failure window that typically creates duplicate writes; verify the repair using repeated and interrupted test runs yielding exactly one committed credit per approved case.

Incorrect Answers

A:

A feedback dataset improves future evaluation; it does not guarantee that the current response is timely or properly permissioned; preserve authorized reviewer labels linked to case versions and turn them into regression examples with data-retention controls; that control is assessed by regression pass rate on previously corrected clause-omission cases; the incident instead requires repeated and interrupted test runs yielding exactly one committed credit per approved case.

C:

Traceability explains decisions after the fact; it is not a replacement for preventing unauthorized source access; correlation IDs, source versions, and stage logs establish end-to-end lineage across extraction, retrieval, inference, and review; that control is assessed by fraction of sampled outputs reproducibly linked to input, model configuration, and reviewer disposition; the incident instead requires repeated and interrupted test runs yielding exactly one committed credit per approved case.

D:

Queueing absorbs bursts but does not improve a weak answer rubric or justify a higher-privilege tool set; decouple intake from processing through durable queues, bounded workers, and status callbacks rather than keeping user connections open; that control is assessed by job completion rate and queue age during the peak intake window; the incident instead requires repeated and interrupted test runs yielding exactly one committed credit per approved case.

Claude Certified Architect - Professional (CCAR-P) Practice Tests