A plain language model can only produce text. An AI agent goes further: it decides what to do next, calls tools such as search, code execution, databases and APIs, looks at the results, and keeps going until the task is done. This module covers function calling, the agent loop, and how agents plan, remember and recover from mistakes.
You'll build agents with popular frameworks, then look at multi-agent systems, the Model Context Protocol (MCP) for connecting tools, and the practical side of making agents reliable: guardrails, evaluation, cost control and human-in-the-loop design.
These courses are long, so treat them as a reference: watch the parts you need, then come back for the rest. When you're done, continue on to Fine-Tuning & Evaluation.
What an AI agent is and how it differs from a plain chatbot: the core components of an agent, and the kinds of tasks where agents work well.
How agents call external functions and APIs: defining tools, letting the model choose which to call, and feeding the results back into the loop.
A full freeCodeCamp course on LangGraph: graph state, nodes and edges, looping workflows, and hands-on agents such as a RAG agent and a document drafter.
Splitting work across several specialised agents, how they communicate and hand tasks off, and when a multi-agent setup is worth the extra complexity.
The standard protocols for connecting agents to tools and to each other, including the Model Context Protocol (MCP), Agent2Agent (A2A) and NLWeb.