CrewAI is an open-source, code-first Python framework for building teams ("crews") of role-playing AI agents that collaborate on a task. In this course you scaffold a crew from one command, learn how agents, tasks and YAML fit together, run it to watch the agents reason, then level up: real tools, your own LLM, the Crews-vs-Flows choice, and the optional hosted AMP platform for a visual editor, GitHub deploy and run monitoring. You write a little Python — the payoff is real, owned agent code rather than boxes on a canvas.
Build your first crew
Work like a pro
A CrewAI crew where a "strategist" picks an angle, a "writer" drafts, and an "editor" tightens it — turning one topic into a finished post through agent hand-offs.
A repeatable pipeline that drafts on-brand content from a single prompt.
Scaffold a crew with strategist/writer/editor roles → give the writer a web-search tool → run it on a topic.
A CrewAI crew of three agents — a "searcher", a "summarizer", and a "critic" — divides a literature review: one finds papers, one extracts findings, one flags weak evidence, producing a structured draft.
A first-pass review draft assembled by agents collaborating, ready for you to verify and refine.
pip install crewai → crewai create crew lit_review → define researcher/writer/critic agents in the YAML → crewai run.