CCAR-P Learning Guide

CCAR-P Learning Guide

Thirty-eight short lessons, one for each task statement in the CCAR-P exam guide, grouped by the seven exam domains. Each lesson explains the decisions the task asks you to make, the traps that catch candidates, and where to read more in Anthropic's documentation.

lessons complete
Domain 117% of the exam

Solution Design & Architecture

Translating business problems into Claude-based solutions: choosing between workflow, agentic and augmented-LLM patterns, designing multi-agent orchestration, and tying the design to business value and SLAs.

6 lessons
Domain 213% of the exam

Claude Models, Prompting & Context Engineering

Selecting models on their trade-offs, designing system prompts and guardrails, applying few-shot and chain-of-thought techniques, and managing context and token usage with caching, compaction and Skills.

5 lessons
Domain 319% of the exam

Integration

Connecting Claude to enterprise systems safely: least-privilege tool sets, authentication and authorization, observability at scale, RAG pipeline design, and choosing between MCP, API/CLI and cloud delivery routes.

8 lessons
Domain 416% of the exam

Evaluation, Testing & Optimization

Proving a system works: defining metrics, building golden datasets and mixed-method graders, running sound A/B and shadow tests, diagnosing failures, and optimizing token usage, latency and cost.

6 lessons
Domain 514% of the exam

Governance, Safety & Risk Management

Keeping systems safe and compliant: layered guardrails, prompt-injection defence, failing closed, human-in-the-loop placement, GDPR, HIPAA and FedRAMP obligations, and fairness and transparency.

5 lessons
Domain 614% of the exam

Stakeholder Communication & Lifecycle Management

Working with the people around the system: structured discovery, presenting trade-offs so stakeholders can act, SLAs and feedback loops, documentation that survives handoff, and the lifecycle from design to iteration.

5 lessons
Domain 77% of the exam

Developer Productivity & Operational Enablement

Enabling engineering teams: configuring Claude Code with CLAUDE.md, Skills, hooks, subagents and permission modes, rolling out AI tooling well, and supporting debugging and operations.

3 lessons

Every lesson ends with a short exercise set in one of four fictional companies. Their facts are on the case studies page.

Ready to test yourself? The CCAR-P practice tests cover the same seven domains, with a rationale and documentation link for every question.