CCAR-P

Claude Certified Architect - Pro (CCAR-P) practice exams

290 questions in 5 timed practice tests. Each test has about 60 questions and a 120-minute limit. The pass mark is 72%. Results are split across the 7 exam domains.

Exam at a glance

CCAR-P practice exam settings
Practice tests 5
Questions in total 290
Questions per test About 60
Time limit per test 120 minutes
Pass mark 72%
Question types Single answer and multiple answer (30 multiple-answer questions)
Level Professional
Category AI & Machine Learning

These are the settings of our practice tests. Check the official Anthropic exam page for the current exam format before you book.

Domains and weights

CCAR-P exam domains, weights and question counts
Domain Weight Weight bar Questions in our bank
Solution Design & Architecture 17% 49
Claude Models, Prompting & Context Engineering 13% 38
Integration 19% 55
Evaluation, Testing & Optimization 16% 46
Governance, Safety & Risk Management 14% 41
Stakeholder Communication & Lifecycle Management 14% 41
Developer Productivity & Operational Enablement 7% 20

Sample questions

Question 1 of 3

A due-diligence platform must research 40 companies per request, each requiring independent web and filings research, then produce a comparative report. A single-agent prototype takes 50 minutes and frequently exhausts its context window. Which redesign BEST addresses both issues?

  1. A An orchestrator agent that delegates each company to a subagent with its own context window and has subagents return condensed findings for synthesis (correct answer)
  2. B A prompt-chaining workflow that researches the 40 companies strictly one after another, persisting each result to a file before starting the next one
  3. C A single agent that calls the Message Batches API for every research step so that the 40 companies are processed at a discount during off-peak hours
  4. D A single agent with a much larger context window and server-side compaction enabled so it can keep all 40 companies' raw research in one long session
Why: The work is breadth-first and parallelizable, which is where an orchestrator-worker multi-agent design shines: subagents explore in parallel within separate context windows and return compressed results, cutting wall-clock time and protecting the lead agent's context. A bigger context window with compaction still runs serially and keeps the latency problem. Sequential chaining fixes context exhaustion but not the 50-minute runtime. The Batches API is asynchronous with results up to 24 hours later, which worsens interactive latency.

Question 2 of 3

An insurer wants Claude to process first-notice-of-loss claims. Every claim follows the same five steps (extract fields, validate policy, check fraud signals, draft a summary, route to an adjuster queue), auditors require a reproducible execution path, and per-claim latency must be predictable. Which architectural pattern BEST fits these requirements?

  1. A An evaluator-optimizer loop where one Claude call drafts the full claim outcome and a second call critiques and revises it until a quality bar is met
  2. B An autonomous agent given all five capabilities as tools, deciding the order of steps for each claim
  3. C A deterministic prompt-chaining workflow in application code, with a programmatic validation gate between steps and a Claude call per step (correct answer)
  4. D An orchestrator-workers design where a lead Claude call dynamically plans subtasks for each claim and spawns parallel worker calls to execute them
Why: A fixed, well-understood sequence with audit and latency-predictability requirements is the textbook case for a workflow (prompt chaining) whose path is defined in code, with gates between steps. A fully autonomous agent chooses its own path, which undermines reproducibility and predictable latency. Orchestrator-workers suits tasks whose subtasks cannot be known in advance, which is not the case here. An evaluator-optimizer loop adds a variable number of iterations and does not provide the fixed, auditable step sequence required.

Question 3 of 3

A multi-agent research product uses one model for planning and synthesis and runs many subagents that perform simple, high-volume source extraction. Costs are high and latency is acceptable. Which model allocation is MOST cost-effective while protecting quality?

  1. A The smallest available model everywhere, including planning and synthesis, since extraction quality is the only thing that affects the report
  2. B A mid-tier model for the orchestrator and the most capable model for the subagents, because subagents read far more raw source material
  3. C A highly capable model for planning and synthesis, and a small fast model for the extraction subagents, validated by evals (correct answer)
  4. D The most capable model everywhere, since mixing tiers causes inconsistent formatting between agents
Why: Matching model tier to subtask difficulty is a key cost lever: complex planning and synthesis benefit from a capable model, while simple, high-volume extraction can usually run on a small fast model, with evaluations confirming quality. Using the smallest model for planning risks poor decomposition and synthesis. Using the top model everywhere wastes spend on simple work, and formatting consistency is handled by output specifications. Putting the most expensive model on the high-volume, simple tasks inverts the allocation.

What your score report looks like

Example attempt, CCAR-P practice test 3

76%

Passed
Pass mark 72%

Score by exam domain
DomainWeightYour score%
Solution Design & Architecture 17%
88
Claude Models, Prompting & Context Engineering 13%
81
Integration 19%
76
Evaluation, Testing & Optimization 16%
58
Governance, Safety & Risk Management 14%
84
Stakeholder Communication & Lifecycle Management 14%
72

Study next: Evaluation, Testing & Optimization. It is below the pass mark and carries 16% of the exam.

Domains and pass mark from our CCAR-P practice tests. Scores are an example attempt.

Study plans

Two weeks

  1. Day 1: take practice test 1 in exam mode to get a baseline.
  2. Days 2 to 5: domain mode on Integration and Solution Design & Architecture, the two most heavily weighted domains.
  3. Day 6: practice test 2 in exam mode, then the mistakes quiz.
  4. Days 7 to 12: practice tests 3 to 5, one every other day, each followed by its mistakes quiz.
  5. Days 13 and 14: final mode. Book the exam when your readiness stays above the pass mark.

Four weeks

  1. Week 1: one practice test in practice mode, reading every explanation. Note your two lowest domains.
  2. Week 2: domain mode on those two domains, then a second practice test in exam mode.
  3. Week 3: practice tests 3 and 4 in exam mode, with the mistakes and flagged quizzes after each.
  4. Week 4: practice test 5 and final mode until your readiness holds above the pass mark.

Questions about the CCAR-P practice exams

Are these real CCAR-P exam questions?
No. We write every question ourselves and map it to the published CCAR-P exam objectives. We do not use leaked or recalled questions.
Do the time limit and pass mark match the official exam?
They match the settings of our practice tests. Exam providers change their format from time to time, so check the official exam page before you book.
Can I try it before I pay?
Yes. The free mini exam has 15 questions with full explanations and needs no account.
How long do I keep access?
A single-exam purchase gives you 12 months of access. All-Access includes this exam for as long as your subscription runs.
What happens after I finish a test?
You get a score, a pass or fail against the pass mark, a score per domain and every answer with its explanation.

Last updated November 2026. Written and reviewed by the DevOpsCode Lab team.

DevOpsCode Lab is not affiliated with Anthropic. CCAR-P and related names are trademarks of their owners.

Related exams
Exam Vendor Level Questions Minutes Price
CCDV-FClaude Developer - Foundations (CCDV-F) Anthropic Associate 300 120 $12.99
CAFClaude Associate Foundations Anthropic Associate 300 120 $12.99
1Z0-1122-26OCI AI Foundations 1Z0-1122-26 Oracle Foundational 300 90 $9.99
AB-731AI Transformation Leader AB-731 Microsoft Foundational 250 100 $9.99
AB-900Copilot & Agent Admin AB-900 Microsoft Foundational 225 45 $9.99
AI-901Azure AI Fundamentals AI-901 Microsoft Foundational 225 45 $9.99