Skip to main content
You have spent the last few weeks learning to direct AI. You have given it roles, briefed it with context, and shaped its outputs. Every time, you were in the driving seat. You prompted. It responded. You reviewed. This week, you step out of the driving seat. AI agents are the next stage in how you work with AI. Instead of prompting every time, you set up the work once (the task, the trigger, the tools) and the agent runs it repeatedly on your behalf. You shift from doing the work to managing the work.

What Makes an AI Agent

Understand the difference between chatbots, copilots, and agents, and when each one is the right choice

Designing Your AI Agent

Map your current workflows, identify where agents fit, and write agent job descriptions using the what/when/how framework

Building and Testing in Relay

Build a working agent in Relay, test it, give feedback, and iterate until it delivers what you need

Challenge: Build Your First Agent

Put it all together, design and build a working agent for a real task in your business

Where you are in the journey

You started as a chatbot user: ask a question, get an answer. Then you became a copilot operator: build context, give clear briefs, get structured outputs. Now you are becoming an agent manager. The skills you built in Weeks 1 and 2 (role assignment, structured prompting, knowledge bases) are exactly what agents need to run well. You are not starting from scratch. You are building on everything you have already practised.

What you will do this week

You will learn what separates an agent from a chatbot or a copilot. You will design an agent using a job description framework. And you will build and test a working agent in Relay, the platform your cohort has access to. By the end of this week, you will have a clear method for identifying where agents add value, a framework for specifying what they should do, and a working prototype you can refine and expand.

What Makes an AI Agent

Understand the difference between chatbots, copilots, and agents, and when each one is the right choice

Designing Your AI Agent

Map your current workflows, identify where agents fit, and write agent job descriptions using the what/when/how framework

Building and Testing in Relay

Build a working agent in Relay, test it, give feedback, and iterate until it delivers what you need

Challenge: Build Your First Agent

Put it all together, design and build a working agent for a real task in your business

Start with What Makes an AI Agent

Begin by understanding the difference between chatbots, copilots, and agents