Passing exam with CCAR-F latest training materials

Anthropic CCAR-F Exam Overview:

Anthropic CCAR-F Exam Syllabus Topics:

Anthropic Claude Certified Architect - Foundations Sample Questions:

Question 1

Your multi-agent research pipeline crashed after processing12 of 28 documents. The web search agent had identified relevant sources, the document analyzer had partially completed extraction, and the synthesizer had begun pattern identification. You need to resume processing without repeating work or losing fidelity of prior findings. What state management approach best balances information fidelity with context efficiency when restoring agent state?

A. Index all agent outputs in a shared vector store. When resuming, each agent queries the store using semantic search to retrieve relevant prior findings.
B. Have each agent persist a structured export to a known location. On resume, the coordinator loads the manifest and injects relevant state into agent prompts.
C. Persist the coordinator's conversation log containing all task delegations and responses, providing this to agents when resuming.
D. Have each agent maintain its own persistent state file and reload it independently at the start of each session.

Question 2

A developer includes multiple unrelated tasks inside one extremely long prompt. What is the MOST likely outcome?

A. Reduced instruction clarity.
B. Increased prompt clarity.
C. Improved reasoning.
D. Faster inference.

Question 3

You are building developer productivity tools using the Claude Agent SDK. The agent helps engineers explore unfamiliar codebases, understand legacy systems, generate boilerplate code, and automate repetitive tasks. It uses the built-in tools (Read, Write, Bash, Grep, Glob) and integrates with Model Context Protocol (MCP) servers.
During testing, you observe that in extended exploration sessions (30+ minutes), the agent starts giving inconsistent answers about code structure it discussed earlier. Engineers report having to repeat context about modules they've already explored. What's the most effective approach to address this?

A. Have the agent maintain a scratchpad file that records key findings, referencing it for subsequent questions.
B. Implement automatic context clearing every 15 minutes to ensure the agent starts with fresh, uncontaminated context.
C. Create summaries of all source files before exploration begins, loading only these compressed representations into context.
D. Switch to a higher-capacity model tier to provide more context window space for accumulated exploration data.

Question 4

You are building a structured data extraction system using Claude. The system extracts information from unstructured documents, validates the output using JSON schemas, and maintains high accuracy. It must handle edge cases gracefully and integrate with downstream systems.
Your invoice extraction uses tool use with strict JSON schemas. JSON syntax errors never occur, but 12% of extractions fail semantic validation--for example, line-item amounts do not sum to the extracted total, or vendor IDs do not match valid formats. These failures currently route to manual review.
What is the most effective approach to reduce manual-review volume while maintaining accuracy?

A. Implement post-processing logic that automatically corrects common errors, such as recalculating totals from line items when sums do not match.
B. When validation fails, make a follow-up request containing the document, extraction, and validation errors so the model can correct the result.
C. Retry the extraction up to three times when validation fails, accepting the first result that passes validation.
D. Add stricter schema constraints with detailed field descriptions to prevent the model from initially generating invalid values.

Question 5

A customer sends: "This is frustrating. I've explained my issue twice and nothing is being resolved. I want to talk to a real person NOW." The agent has not yet called any tools to investigate their account. What should the agent do?

A. Immediately call escalate_to_human with the conversation history.
B. Briefly explain what the agent can help with and offer to resolve the issue quickly, escalating only if the customer repeats their request.
C. First call get_customer and lookup_order to gather account context, then escalate to a human agent.
D. Acknowledge the frustration and ask one targeted question to understand the specific issue before escalating.

Solutions: