AI glossary
AI agent
An AI agent is a program in which a language model chooses its own next step, calling tools in a loop until the task is done or it is stopped.
A plain model call answers once. An agent is given a goal and tools, and repeats: think, call a tool, read the result, decide what next. A coding agent might read a file, run the tests, edit, and run them again.
The useful part is that the path is not written in advance. The cost is that it is harder to predict, test and bound, so an agent needs limits on steps, spend and what its tools may touch.
Many tasks called "agents" are better as a fixed workflow. If you can write the steps down, write them down and use the model only where judgement is needed.
Learn it hands-on
These free lessons teach it with commands to run in a lab and checked tasks. Read them whole with no account.