AI Agents Fundamentals: How Agents Think, Act and Stay Safe
Learn what an AI agent really is, how models, prompts, tools, memory and the agent loop work, when to use one, and the controls that keep agents safe. No code needed.
What you'll learn
- Foundations
- How agents work
- Judgement and safety
- Final assessment
This is the first course in the "Become an AI agent builder" track. It teaches the ideas behind AI agents in plain language, the same ideas that apply to any agent platform, and lets you try each one hands-on in PGB Agents. You will learn how an agent differs from a chatbot and from simple automation, how a model turns a prompt into a decision, what tools and memory add, how the think, act, observe loop works, when an agent is the wrong choice, and which risks and controls matter most. No coding is needed.
The track continues with "Build Your First Agents on PGB Agents" (course slug build-agents-on-pgb-agents), then "Running Agents Safely and Reliably" (run-agents-safely), and finishes with "Agents for Business Teams" (agents-for-business-teams). If you want a quick tour of the product screens first, take "PGB Agents 101" (pgb-agents-101); it is a good companion to this course.