Learning path · Agents & Orchestration · 65
CrewAI
Multi-agent framework organizing agents as crews with roles, goals, and delegated tasks—emphasizing collaborative role-play.
Why it matters
- Fast to prototype team-like workflows.
- Role prompts can drift without hard graph constraints.
- Compare against LangGraph when reliability dominates.
Key ideas
- Role definitions
- Task delegation
- Sequential and hierarchical crews
Top resources
- 01DocsCrewAI
CrewAI documentation
Why this resource. CrewAI's own model of roles, tasks, and crews.
Covers in this concept
- agents
- tasks
- crews
- 02ArticleAnthropic
How we built our multi-agent research system
Why this resource. Skeptical companion: multi-agent cost and failure modes.
Covers in this concept
- coordination
CrewAI assigns personas—Research Analyst, Editor—and chains their outputs. It excels in demos and content pipelines where flexible collaboration beats rigid graphs. Tradeoffs versus LangGraph: less explicit state-machine control, harder step-level replay. Versus AutoGen: more opinionated structure, less free-form group chat. Productionize with tool schemas, eval harnesses, and caps on delegation depth. Snapshot crew role prompts in git and diff them during incidents; role drift is a common source of sudden quality cliffs.
Updated 2026-08-09 · Full learning path