Topic map

Concepts

How language models workCapabilities and limitsPromptingTokens, cost and effortWhat an agent isGrounding and memory

Safety

Recognizing failureResponsible useVerifying outputsAgent riskGovernance and oversight

Using agents

Delegating to an agentWorking with an assistantChoosing models and toolsDecomposing work

Coding with agents

Running a coding agentDeciding and specifyingPlan, implement, verifyContext engineering for codeVerifying agent workQuality with agentsAgents in a team

Customizing agents

Inspecting what the agent sendsProject instructionsAgent skillsConnecting tools with MCPHooks, permissions, settingsMemory and session context

Building agents

Tool useThe agent loop and harnessAgentic retrieval and memoryDesign patternsEvaluation and testingGuardrails and productionOrchestration and multi-agent

not startedin progressfinished

Six areas, each a column. Arrows run from a prerequisite topic to the topic that builds on it. A topic takes the color of the lessons that cover it. Click a topic for its definition, concepts, and lessons. For what you can do after each course, see the competency map.