Claude Certified Architect, Foundational: 385 Questions

This exam expects you to think like an architect.

The Claude Certified Architect – Foundations exam doesn't reward recall. It puts you inside a production system — mid-incident, mid-design, mid-tradeoff — and asks: what's the right call?

Agentic pipelines. MCP tool integration. Multi-agent orchestration.  They're the substrate of every question.

This course builds the judgment to answer them.

Every practice question is a scenario and not a single theory question and you'll be handed a real-sounding system with real constraints — latency limits, compliance requirements, tool boundaries, failure modes — and asked to make an architectural decision. Then you'll get a full breakdown: why the correct answer holds, and exactly where each distractor breaks down.

That's the skill the exam measures. That's what this course trains.

What you'll walk away with:

• Confidence in scenario-based reasoning, not just content familiarity

• Deep coverage across all five exam domains

• The ability to eliminate wrong answers — not just guess the right one

If you've studied the material and still feel unsure when the question gets complicated, this is where that changes.

About the CCA-F Exam

The CCA-F exam consists of 60 scenario-based multiple-choice questions. Candidates have 120 minutes to complete the exam. The passing score is 720 on a scaled score of 100 to 1,000. The exam is delivered via online proctoring or at a testing centre. Ai

The exam's 60 questions are anchored to 6 production scenario contexts. During the exam, 4 of the 6 are randomly selected — so you must study all 6. Each scenario overlaps multiple domains, so studying one reinforces the others. Medium

More than half the exam (45%) is concentrated in agentic architecture and Claude Code configuration. This is not a prompting fundamentals test — it is a systems design exam. DEV Community

What Makes This Course Different

Most practice resources test recall. This course tests the same thing the real exam tests — architectural judgment under production conditions. Every question is grounded in a realistic failure mode: a customer support agent skipping identity verification, a multi-agent research system missing entire topic areas, a CI/CD pipeline hanging because Claude Code is waiting for interactive input. You practice deciding correctly in those situations, not memorising parameter names.

Domains Covered

The course is organised around all 5 official CCA-F exam domains.

Domain 1 — Agentic Architecture and Orchestration (27%) Agentic loop control flow using stop_reason. Multi-agent coordinator and subagent patterns. Subagent spawning via the Task tool. Agent SDK hooks including PostToolUse for data normalisation and compliance enforcement. Task decomposition strategies — prompt chaining versus dynamic adaptive decomposition. Session resumption and fork_session for parallel exploration.

Domain 2 — Tool Design and MCP Integration (18%) Writing tool descriptions that eliminate selection ambiguity. Structured error responses with errorCategory and isRetryable flags. Scoping tool access per agent role. MCP server configuration at project versus user level with environment variable expansion. Built-in tool selection — when to use Grep versus Glob versus Read versus Edit.

Domain 3 — Claude Code Configuration and Workflows (20%) CLAUDE configuration hierarchy across user, project, and directory levels. Path-specific rules in .claude/rules/ with YAML frontmatter glob patterns. Custom slash commands in .claude/commands/ and skills in .claude/skills/ with context fork and allowed-tools. Plan mode versus direct execution — how to choose. Integrating Claude Code into CI/CD pipelines with the -p flag and --output-format json.

Domain 4 — Prompt Engineering and Structured Output (20%) Designing prompts with explicit criteria to reduce false positives. Few-shot prompting for output consistency and ambiguous-case handling. Enforcing structured output with tool_use, JSON schemas, and tool_choice modes. Validation and retry loops for extraction quality. Batch processing with the Message Batches API — when to use it and when not to. Multi-pass and multi-instance review architectures.

Domain 5 — Context Management and Reliability (15%) Managing context across long conversations — progressive summarisation risks, lost-in-the-middle effects, and verbose tool output trimming. Escalation decision-making with explicit criteria and few-shot examples. Structured error propagation across multi-agent systems. Context management in large codebase exploration using scratchpad files and subagent delegation. Human review workflows with confidence scoring and stratified sampling.

All 6 Production Scenarios Covered

The real exam randomly selects 4 of the 6 official scenarios during your sitting. This course covers all 6, so you are prepared regardless of which combination appears. Questions in this course are mapped to these scenarios so you experience the same cross-domain reasoning the real exam demands.

The exam uses scenario-based questions. Each scenario presents a realistic production context that frames a set of questions. During the exam, 4 scenarios will be presented and picked at random from the full set of the 6 scenarios below.

Scenario 1: Customer Support Resolution Agent

You are building a customer support resolution agent using the Claude Agent SDK. The agent handles high-ambiguity requests like returns, billing disputes, and account issues. It has access to your backend systems through custom Model Context Protocol (MCP) tools (get_customer, lookup_order, process_refund, escalate_to_human). Your target is 80%+ fi rst-contact resolution while knowing when to escalate.

Primary domains: Agentic Architecture & Orchestration, Tool Design & MCP Integration, Context Management & Reliability

Scenario 2: Code Generation with Claude Code

You are using Claude Code to accelerate software development. Your team uses it for code generation, refactoring, debugging, and documentation. You need to integrate it into your development workfl ow with custom slash commands, CLAUDE confi gurations, and understand when to use plan mode vs direct execution.

Primary domains: Claude Code Confi guration & Workfl ows, Context Management & Reliability

Scenario 3: Multi-Agent Research System

You are building a multi-agent research system using the Claude Agent SDK. A coordinator agent delegates to specialized subagents: one searches the web, one analyzes documents, one synthesizes fi ndings, and one generates reports. The system researches topics and produces comprehensive, cited reports.

Primary domains: Agentic Architecture & Orchestration, Tool Design & MCP Integration, Context Management & Reliability

Scenario 4: Developer Productivity with Claude

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.

Primary domains: Tool Design & MCP Integration, Claude Code Confi guration & Workfl ows, Agentic Architecture & Orchestration

Scenario 5: Claude Code for Continuous Integration

You are integrating Claude Code into your Continuous Integration/Continuous Deployment (CI/CD) pipeline. The system runs automated code reviews, generates test cases, and provides feedback on pull requests. You need to design prompts that provide actionable feedback and minimize false positives.

Primary domains: Claude Code Confi guration & Workfl ows, Prompt Engineering & Structured Output

Scenario 6: Structured Data Extraction

You are building a structured data extraction system using Claude. The system extracts information from unstructured documents, validates the output using JavaScript Object Notation (JSON) schemas, and maintains high accuracy. It must handle edge cases gracefully and integrate with downstream systems.

Primary domains: Prompt Engineering & Structured Output, Context Management & Reliability

Who This Course Is For

• Solution architects and engineers building production Claude applications

• Technical professionals at Anthropic partner organisations preparing for the CCA-F

• Developers with hands-on Claude API, Claude Code, or MCP experience who want to validate their skills

• Consultants and delivery leads at firms where the CCA-F is becoming a baseline expectation

• Anyone with at least 6 months of hands-on production experience with the Claude API, Claude Code, and MCP Medium who is ready to sit the exam

What You Will Get

• Scenario-based multiple-choice questions mirroring the real exam format and difficulty

• Full coverage of all 5 domains and all 6 official production scenarios

• Detailed explanations for every answer — correct answers explained and each distractor examined

• Domain-mapped organisation so you can study targeted areas or simulate full exam conditions

• Questions calibrated to the 301-level architectural judgment the exam is designed to test