AI agent orchestration is coordinating the steps, tools, models and people an AI agent relies on — deciding what runs when, passing data between steps and enforcing the controls around them.
Also called: agent orchestration, LLM orchestration, multi-agent orchestration, AI workflow orchestration
A useful AI agent is rarely one model call. It retrieves context, calls a model, checks the output, calls a system, waits for a person and continues. Orchestration is the layer that sequences those steps, handles loops, branches, retries and failures, and passes typed data from one step to the next.
In an enterprise, orchestration is also where control lives: which model each step may use, where a guardrail must run, where a human must approve, and what gets logged. ATX Agent Hub expresses orchestration as visual workflows, with each step versioned, tested and traced.
This site needs JavaScript for the full experience. Every page above is reachable without it.