Curriculum
All 30 lessons
The full path to the exam — 5 domains, weighted by their share of the test. No filler; each lesson lands a concrete idea you'll be tested on.
- Agent vs. WorkflowWhen to give Claude autonomy vs. when to pin a fixed sequence.6 min
- Orchestrator-Worker PatternOne Claude routes subtasks to specialized subagents.8 min
- Evaluator–Optimizer LoopUse Claude to grade Claude's own output and iterate.7 min
- Parallel Fan-OutMap-reduce over Claude calls when the work shards cleanly.6 min
- Agent State & MemoryPersisting decisions and intermediate results across turns.9 min
- Human-in-the-LoopWhere to surface confirmation prompts without breaking flow.5 min
- Evaluating Agent QualityGolden tasks, success rates, and step-level traces.7 min
- Tool Schema FundamentalsDesigning input schemas Claude reliably populates.6 min
- Tool Error ContractsReturn errors that teach Claude how to recover.5 min
- MCP OverviewWhat the Model Context Protocol is and when to use it.8 min
- MCP Resources vs. ToolsWhen to expose data as a resource vs. a tool call.6 min
- Testing Tools in IsolationRun tools end-to-end without invoking Claude.5 min
- Claude Code: What It IsAnthropic's official agentic CLI for software work.5 min
- settings.json ConfigurationWhere the harness reads permissions, hooks, env, and model.7 min
- HooksShell commands that fire on Claude Code events.8 min
- Slash CommandsReusable named workflows the user invokes with /name.6 min
- SkillsModular capability packs with their own prompts and resources.7 min
- Claude Agent SDKBuilding custom agents on the same primitives as Claude Code.9 min
- System PromptsWhere to put role, constraints, and output contract.6 min
- Structured OutputJSON mode, schema-guided generation, and post-validation.7 min
- Few-Shot ExamplesHow many examples, in what order, of what shape.6 min
- Prompt Caching5-minute server-side cache that cuts tokens by 90% on hits.8 min
- Extended ThinkingWhen to enable thinking budgets and what they cost.7 min
- Iterating Prompts Like CodeVersion, test, and review prompts.5 min
- Token BudgetsWorking within context windows without truncating critical state.6 min
- Context CompactionSummarize old turns to keep recent detail.8 min
- Retries & BackoffIdempotency and exponential backoff for API errors.5 min
- ObservabilityLogging traces, costs, and quality signals.7 min
- GuardrailsInput/output filters that catch policy and PII issues.6 min
- Safe RolloutsFeature flags, shadow runs, and gradual ramps.7 min