AI Agent Essentials

You've heard a lot about AI agents. Maybe you've even seen one in action — an AI that books meetings, processes invoices, or drafts follow-ups without anyone clicking a button. It looks impressive. And a little unsettling.

Here's the thing nobody tells you: the people who get the most out of AI agents aren't the ones who hand everything over to the technology. They're the ones who understand exactly what the agent is doing — and know how to keep it in their control.

That's what this class is for.

In four sessions, you'll go from "I've heard of agents" to designing a real one. You'll learn how AI agents actually work — the six components, the three levels, the difference between an agent that amplifies your judgment and one that quietly substitutes for it. You'll design the instruction set that governs your agent's behavior, set the permission boundaries that limit its blast radius, and build the safety architecture that keeps a human in the loop at the moments that count.

And you'll do it with a real workflow from your own work. Not a hypothetical. Not a textbook example. Your work.

By the end of Session 4, you'll submit a complete Agent Design Plan to your AI capstone evaluator — a five-component document specific enough to implement, review, or hand to a developer. Pass the badge rubric and you earn the AI Agent Essential Badge.

What you'll build:

• A workflow map with goal, steps, and human review gate

• A three-layer instruction architecture with system prompt and memory plan

• A permission scope table and guardrail design

• A full Agent Design Plan — your professional governance document for AI deployment

Who this is for: Working professionals ready to move beyond using AI tools to governing them. No coding required. Experience with AI assistants helpful.

AI is not replacing you. It's amplifying you. This class is where that stops being a slogan and starts being a skill.