AI agent glossary
Workflow vs. agent
A workflow follows steps you define in code, while an agent lets the model decide its own steps; each suits different kinds of tasks.
In a workflow, you decide the sequence: call the model to classify the request, then call it again to draft a reply, then run a check. The model fills in each step, but the path is fixed. In an agent, you give the model a goal and tools and let it choose the path.
Workflows are cheaper, faster, easier to test, and more predictable. They are the right choice when the steps are known in advance. Agents are more flexible and handle open-ended tasks where the right next step depends on what was just discovered.
Many real systems mix the two: a fixed workflow at the top level, with an agent handling one open-ended step inside it. Start with the simplest design that works, and add autonomy only where it earns its cost.
Translating every new help article into five languages is a workflow. Investigating why signups dropped last week, where each finding changes what to check next, suits an agent.
Learn more in the course
Related terms
Go from definitions to doing
Learn to use and build AI agents in 65 practical lessons.