CCAR-P Practice Test 13 — 15 Questions

Question 1

A platform team maintains a shared Claude API integration used by multiple product squads. Squads frequently push prompt changes that break downstream features, and nobody can tell which prompt version produced a given output in production. The team wants every API call to be traceable to an exact prompt revision and wants to gate prompt changes behind review. Which approach best satisfies both requirements?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Store prompts in a Git repository, reference each revision by commit SHA when calling the Messages API, and require pull-request review before merging prompt changes.. Prompts are code and should be treated as such. Git provides immutable commit SHAs that can be logged with each Messages API call, giving exact traceability from a production output back to the prompt revision that generated it. Pull-request review adds a human gate so breaking changes are caught before they reach the shared integration, satisfying the change-control requirement.

Answer

Enable extended thinking on every request and rely on the model's reasoning output to reconstruct which prompt was used.

Answer

Add a timestamp field to every API request and correlate logs with the deployment time of the calling service.

Answer

Have each squad maintain its own prompt copy in a shared wiki page and record the editor's name in the page history.

Question 2

An organization runs a nightly batch job that uses the Claude Messages API to classify support tickets. The job currently processes tickets one at a time, taking several hours and occasionally exceeding the nightly window. The team wants to cut wall-clock time substantially without exceeding their rate limits or degrading classification quality. Which change is most effective?

Accepted Answer

Submit the batch via the Message Batches API, which processes requests asynchronously at a lower cost and returns results without holding a synchronous connection.. The runtime problem is a throughput problem, not a model-quality problem. The Message Batches API is purpose-built for high-volume, non-interactive work, processing requests asynchronously and at lower cost than synchronous calls. For a nightly classification job that does not need real-time responses, batching removes the sequential bottleneck and shortens wall-clock time while staying within rate limits.

Answer

Split the tickets across multiple API keys belonging to different team members to multiply the effective rate limit.

Answer

Increase the max_tokens value on each request so the model has more room to reason before classifying.

Answer

Lower the temperature to zero and retry any borderline classifications until the model returns a confident label.

Question 3

An engineering organization wants to raise developer productivity across teams building on Claude. Leadership asks the platform team to identify interventions that reduce repeated manual work and shorten the feedback loop for prompt and integration changes. (Choose two.)

Accepted Answer

Provide a reusable internal library that wraps common Claude API calls, including retry, streaming, and token accounting, so teams stop reimplementing the same plumbing.. The two interventions that directly reduce repeated work and tighten feedback are a shared client library that absorbs common plumbing and a centralized evaluation harness that reports quality and latency deltas on every pull request. Together they remove duplicated effort and surface regressions early, letting teams iterate faster while keeping results comparable across services.

Answer

Mandate that all prompt changes be approved by a central architecture board that meets twice monthly.

Answer

Require every team to freeze its prompt text for a full quarter so that results remain comparable across services.

Answer

Instruct each team to build its own bespoke evaluation scripts and dashboards so that tooling matches local needs exactly.

Question 4

An insurance company is preparing an AI risk register for its Claude-based claims triage system. The risk team must document controls that reduce the chance of biased or inconsistent decisions affecting policyholders. (Choose two.)

Accepted Answer

Define and version the decision criteria and prompts used for triage, and re-validate them against a representative dataset whenever they change.. Fairness controls require both a defined, versioned decision logic that is re-validated against representative data and a human appeal path that catches harmful outcomes the model produces. Together they create a preventive and a corrective layer. Throughput, token budgets, and model size describe performance or capability, not equity, and cannot substantiate a claim that biased or inconsistent decisions have been controlled.

Answer

Collect aggregate throughput metrics showing how many claims the system processes per hour during peak periods.

Answer

Raise the model's max_tokens setting so the triage system can produce longer, more detailed justifications for each decision.

Answer

Switch to the largest available Claude model on the assumption that greater capability automatically eliminates biased outputs.

Question 5

An agent orchestrator delegates work to three specialist sub-agents: a 'search' agent that returns ranked documents, an 'extract' agent that pulls structured fields from those documents, and a 'verify' agent that checks extracted fields against source text. During evaluation you find that verify frequently approves fields that extract hallucinated, because verify receives only the extracted JSON, not the source passages. Which change to the orchestration contract most directly fixes this?

Accepted Answer

Have the extract agent include, for each field, the source span it was derived from, and pass those spans to the verify agent alongside the extracted JSON.. The verifier approves hallucinations because it is asked to judge extracted values without access to the text they were derived from. Propagating per-field source spans gives the verifier the evidence needed to confirm or refute each value against its origin. This is a contract change between extract and verify, ensuring provenance flows with the data rather than being reconstructed or guessed downstream.

Answer

Route all extracted fields back through the search agent so it can re-rank the documents before verification.

Answer

Increase the verify agent's temperature so it is more likely to question the extracted fields it receives.

Answer

Add a fourth 'audit' agent that independently re-runs the extract agent on the same documents and compares the two JSON outputs.

Question 6

You are designing a Claude agent that must complete a multi-hour research task spanning dozens of tool calls, and the transcript will eventually exceed the model's context window. You want the agent to keep making correct decisions without losing critical earlier findings. Which TWO architectural strategies best preserve decision quality across the compaction boundary? (Choose two.)

Accepted Answer

Maintain an external structured scratchpad of confirmed findings, open questions, and decisions, and inject a curated summary of it into each new context window.. Durable continuity comes from moving authoritative state outside the transcript. A structured scratchpad preserves confirmed findings and open questions, while a persisted task state object lets the agent rehydrate deterministically at each boundary. Together they let compaction discard raw turns safely. Token limits and internal reasoning do not expand context or survive eviction.

Answer

Summarize and discard the oldest transcript turns once a threshold is reached, retaining only the most recent turns verbatim.

Answer

Enable extended thinking on every turn so the model can reason through the full history internally without needing external state.

Answer

Raise the max_tokens parameter on every request so the model can hold more of the transcript in a single response.

Question 7

A support engineering team wants new hires to become productive with the company's internal Claude-powered assistant quickly. They need a single place where engineers can discover approved prompt patterns, see working request examples, and read guidance on handling tool-use results. Which artifact best serves this enablement goal?

Accepted Answer

A curated internal developer portal page containing vetted prompt templates, runnable request and response examples, and tool-use handling guidance, kept current by the platform team.. Enablement content works when it is curated, current, and executable. A maintained portal lets new hires discover vetted prompt templates, copy working request examples, and read guidance on tool-use results in one place. Ownership by the platform team keeps it aligned with the evolving API and internal conventions, which directly reduces time to first productive contribution.

Answer

A generated API reference produced directly from the vendor's public documentation and mirrored nightly into the internal wiki.

Answer

A read-only archive of the last six months of Slack messages from the team's #claude-help channel, searchable by keyword.

Answer

A shared drive folder of presentation decks from past architecture reviews, organized by fiscal quarter.

Question 8

You operate a long-running research agent that maintains a scratchpad of findings across many turns. You notice that as the scratchpad grows, the agent begins ignoring recent tool results and repeating earlier conclusions. You cannot increase the context window. Which intervention most directly addresses the root cause of the recency failure?

Accepted Answer

Restructure the scratchpad into a fixed-size rolling summary that is regenerated each turn, with the most recent tool results kept verbatim in a dedicated, clearly delimited section.. The agent is not forgetting recent results so much as losing them in an ever-growing block of text where early material dominates attention. Bounding the scratchpad with a regenerated rolling summary keeps total size stable, and reserving a clearly delimited section for the newest tool results puts current evidence in a position the model reliably attends to. This fixes the structural cause rather than the sampling or storage symptoms.

Answer

Move the scratchpad out of the prompt entirely and store it in an external vector database, retrieving relevant entries only when the agent explicitly asks.

Answer

Append every tool result verbatim to the scratchpad so no information is lost, and instruct the agent to re-read the entire scratchpad before each decision.

Answer

Lower the model's temperature to zero so the agent becomes more deterministic and less likely to drift from the current evidence.

Question 9

You are operating a Claude-based support agent that must follow a strict refund policy: refunds over $500 require a manager approval code that is only obtainable through a separate internal API. The agent has access to a `get_manager_code` tool, but in production it occasionally issues refunds above $500 without calling the tool. Which architectural change most reliably prevents this?

Accepted Answer

Add a pre-tool-use hook that inspects the refund amount and blocks any refund tool call above $500 unless a valid manager code is present in the session state.. Hard business constraints must be enforced deterministically outside the model. A pre-tool-use hook inspects the pending refund call and rejects any amount above $500 that lacks a validated manager code, making the violation impossible rather than improbable. Prompting, temperature, and token limits only shift probabilities and cannot guarantee compliance with a policy that carries financial or legal consequences.

Answer

Expand the system prompt to describe the $500 threshold in bold and add two few-shot examples of correct refund handling.

Answer

Increase the max_tokens setting so the agent has more room to reason about whether approval is needed.

Answer

Lower the model temperature to 0 so the agent produces deterministic decisions.

Question 10

A media company wants a lightweight, recurring review of its Claude-powered content moderation assistant. The team has no dedicated compliance staff and wants the cheapest process that still produces defensible evidence of oversight. Which approach fits best?

Accepted Answer

Adopt a scheduled sampling review where a rotating staff member scores a fixed sample of outputs against written criteria and records the results.. Defensible oversight means showing that a named person reviewed outputs against written criteria on a defined schedule and recorded the outcome. Scheduled sampling achieves exactly that with minimal tooling and staffing, and the recorded scores form a durable evidence trail. Vendor documentation and complaint trends are supporting signals, not substitutes for local, documented human review.

Answer

Commission an annual third-party audit with formal attestation and a published report.

Answer

Track the volume of user complaints and treat a stable or declining trend as sufficient evidence of adequate oversight.

Answer

Rely on the vendor's published safety evaluations and model cards as the primary evidence of the assistant's suitability.

Question 11

A software company is using Claude to generate code snippets for internal projects. The security team is concerned that the model might inadvertently suggest code with known vulnerabilities. Which governance control should be implemented to best mitigate this risk?

Accepted Answer

Integrate a static application security testing (SAST) tool into the CI/CD pipeline to scan all AI-generated code before merging.. Integrating SAST into the CI/CD pipeline is the most effective control because it automatically scans all AI-generated code for known vulnerabilities before merging. This provides a consistent, scalable, and early detection mechanism. Manual review and fine-tuning are helpful but less reliable, and restricting scope does not address the core risk of vulnerable code.

Answer

Require developers to manually review all AI-generated code for security issues.

Answer

Restrict Claude's access to only generate code for non-critical components.

Answer

Fine-tune Claude on a dataset of secure code examples to reduce the likelihood of generating vulnerable code.

Question 12

A Claude agent performs a multi-step deployment task. Step 3 calls a `deploy_service` tool that returns success, but the subsequent verification step fails because the service is not yet healthy. The agent currently treats any tool success as completion and ends the workflow. Which change best addresses this?

Accepted Answer

Add a post-tool-use hook that polls the service health endpoint and, if unhealthy, returns a structured error to the model so it can retry or escalate.. The agent conflates tool acceptance with desired end state, so verification must be moved into a deterministic post-tool-use hook that polls health and returns a structured error when the service is unhealthy. That lets the model retry or escalate instead of ending the workflow prematurely. Prompting, return-type changes, and token limits all leave the false-success gap intact.

Answer

Instruct the model in the system prompt to always wait 60 seconds and re-check health after every deploy.

Answer

Increase the agent's max_tokens so it has more room to notice the verification failure in its reasoning.

Answer

Change the `deploy_service` tool to return a boolean instead of a status string so the agent can parse it more easily.

Question 13

An engineer is onboarding to a codebase that calls Claude and needs to understand, at a glance, which model, temperature, and max_tokens values a given feature uses. The team wants this discoverable without reading application source. What is the most effective practice?

Accepted Answer

Store model and sampling parameters in a versioned configuration file that the application loads at startup.. Parameters that drive model choice, cost, and behavior belong in a versioned configuration file rather than scattered through code or held in memory. This makes them discoverable at a glance, reviewable in pull requests, and changeable without a code deploy. Comments, debug logs, and tribal knowledge each fail at least one of those properties, which is why configuration-as-code is the standard practice for developer enablement.

Answer

Rely on code comments near each API call to document the chosen parameters.

Answer

Ask each engineer to memorize the parameters for the features they own.

Answer

Log the parameters at debug level so they appear in application logs when needed.

Question 14

Midway through a Claude integration project, a business stakeholder asks you to add a feature that would let end users upload arbitrary documents and have Claude answer questions about them. The feature is not in the signed statement of work, and the delivery team is two weeks from the first production milestone. What should you do first?

Accepted Answer

Perform a lightweight impact assessment covering effort, risk, cost, and milestone impact, then present options with a recommendation to the sponsor and stakeholder.. Scope changes mid-project are normal, and the disciplined response is a fast, lightweight impact assessment that converts an informal ask into a decision the sponsor can make with eyes open. Presenting effort, risk, cost, and milestone impact alongside options respects both the signed commitment and the stakeholder's underlying need.

Answer

Escalate the request to the project management office and let them decide whether to approve the change.

Answer

Accept the request and absorb the work into the current sprint so the stakeholder sees responsiveness.

Answer

Decline the request and tell the stakeholder to raise it at the next quarterly planning cycle.

Question 15

A product owner tells you the Claude assistant 'feels slow' for end users and wants to know what to tell the executive sponsor. You have measured that time-to-first-token is 400 ms but full responses take 9 seconds because outputs are long. Which communication is most accurate and useful for the sponsor?

Accepted Answer

Explain that perceived slowness comes from long output generation, not connection setup, and propose streaming tokens to the UI plus asking Claude for more concise answers.. Good stakeholder communication distinguishes between what users perceive and what the system actually measures. Fast time-to-first-token with slow completion is the signature of long outputs, so the honest message names that cause and pairs it with remedies that match it: streaming to improve perceived responsiveness and output-length guidance to reduce real duration. Blaming infrastructure, swapping models without evidence, or declaring the problem nonexistent all misrepresent the measurements.

Answer

Tell the sponsor that latency is an infrastructure problem owned by the platform team and hand off the ticket.

Answer

Report that latency is within acceptable bounds and no action is needed.

Answer

Recommend switching to a different Claude model tier immediately to resolve the speed complaint.