Kairox Arch Prep
CCA-F PREP · KAIROX ARCH PREP

Prepare for the Claude Certified Architect exam.

Original scenario questions, a timed practice exam weighted like the real blueprint, and plain explanations of every domain.

01 / START HERE

Not sure where you stand? The diagnostic takes about fifteen minutes: two questions per domain, a score for each, and an explanation for every answer.

Exam blueprint: share of the Claude Certified Architect – Foundations exam by domain1. Agentic architecture27%≈16 of 602. Claude Code20%≈12 of 603. Prompts & output20%≈12 of 604. Tools & MCP18%≈11 of 605. Context & reliability15%≈9 of 60
The exam blueprint. Our practice exam draws questions in the same proportions, so your per-domain score lines up with how the exam is weighted. The largest share, agentic architecture, deserves the most study time.

Study the five domains

The exam, in brief

The Claude Certified Architect – Foundations exam tests judgment about trade-offs when building with Claude: the Claude API, the Agent SDK, Claude Code and the Model Context Protocol. You get 120 minutes for scenario-based multiple-choice and multiple-response questions, scored on a scaled 100–1,000 range with a pass mark of 720. Registration is currently limited to people at Claude Partner Network organizations, and the list price is $125.

Try one now

Sample question 1

Scenario: Customer support resolution agent

A support agent built on the Claude API calls tools to look up orders and issue refunds. In testing, the loop sometimes keeps calling tools for 40+ turns when a customer record is malformed, running up cost. What is the most appropriate first change?

  1. A. Raise max_tokens so Claude can complete its reasoning in fewer, longer turns per request
  2. B. Add a system prompt rule telling Claude to stop after a few unsuccessful tool calls in a row, then apologise
  3. C. Switch to a larger model, which is less likely to repeat the same failing tool call
  4. D. Enforce a maximum iteration count in the orchestrator and escalate to a human when it is reached
Show answer and explanation

Answer: D. A bounded loop is a basic production control: an iteration (or cost) cap in code makes runaway behaviour impossible whatever the model does, and escalating turns the failure into a handled case. A prompt rule is guidance, not enforcement. max_tokens limits one response, not the number of turns. A bigger model may loop less but guarantees nothing and costs more.

There are 71 original questions in our bank, each with an explanation of why the right answer is right and why the others fail.

Honest practice

Every question here was written for this site from the public blueprint and Anthropic's documentation. None is a real exam item, and we do not sell or link to "exam dumps". Facts about fees and rules are checked against Anthropic's certification FAQ and dated on each page. Kairox Arch Prep is independent and not affiliated with Anthropic.