On its own, a language model can only produce text. Tools let it do more: search the web, query a database, run code, or call an API. With function calling, the model decides when a tool is needed and what to send it, and your code runs the tool and returns the result.
You'll start with how tool use and function calling work, including how the model picks a tool and fills in its inputs. Then you'll see how agents chain these calls together, deciding on a step, using a tool, checking the result, and repeating until the task is done.
Tools and agents can do real work, so you'll want to know whether they do it well. That's the focus of Evaluating & Iterating Prompts, the next module in this course.
How language models use external tools such as APIs, databases, web search, and code execution through function calling.
How a language model can decide to call an outside function, and what that means for what it can do.
How tool calling lets AI agents browse the web, call APIs, and automate tasks.
The basics of AI agents and agentic workflows, with simple, practical examples.