Agentic AI
BasicAdvancedAI agents that decide and act — the loop, tools, memory, planning, and what it takes to make one reliable, observable and safe.
6
modules
14
lessons
4
runnable exercises
Sign in to begin →Sign in, then straight into the first lesson.
Course contents
01What an agent isBasic
What is an agent?
The difference between a model that answers and an agent that acts.
The agent loop
Perceive, decide, act — the cycle at the heart of every agent.
02Giving an agent capabilityBasic
Tools and actions
How an agent reaches beyond text to search, calculate and fetch.
Memory and state
How agents carry information across steps and sessions.
03Planning and reasoningBasic
Planning and decomposition
Breaking a goal into steps, and when a plan helps rather than hinders.
Reasoning and acting
Making the agent think out loud, and why that changes what it can do.
04Making agents reliableAdvanced
Failure and recovery
Tools fail; the difference between a robust agent and a stuck one is what happens next.
Budgets and termination
Limits the model cannot talk its way past.
Human in the loop
Deciding which actions an agent may take alone.
05Agents in a systemAdvanced
Multiple agents
Splitting work across agents, and the cost nobody counts.
Observability and tracing
You cannot debug what you did not record.
06Agents and riskAdvanced
Agent security
What changes when a system that can be manipulated is also allowed to act.
Evaluating agents
Scoring a process, not just an answer.
When not to build an agent
The most useful judgement in the course.
Ready when you are.
Opening a lesson takes an account. Nothing to install.