AI agent glossary
Orchestrator-worker pattern
A multi-agent design where a lead agent breaks a task into pieces, hands each to a worker agent, and combines their results.
In the orchestrator-worker pattern, one agent acts as the coordinator. It understands the overall goal, decides how to divide the work, writes a clear brief for each worker, and assembles what comes back into a final result.
Workers are usually focused and short-lived. Each gets just the context it needs for its piece, does the work, and returns a concise summary. Because workers can run in parallel, this pattern is well suited to broad research, large code changes across many files, or processing many documents.
The quality of the briefs decides the quality of the result. Vague delegation leads to overlapping or off-target work, so the orchestrator should state each worker's goal, scope, and expected output format explicitly.
Researching a market, an orchestrator assigns one worker to competitors, one to pricing, and one to regulations, then merges their findings into a single report.
Learn more in the course
Related terms
Go from definitions to doing
Learn to use and build AI agents in 65 practical lessons.