Agents & tools
Agent Orchestration
The coordination layer that decides which agent or tool runs next, manages state across steps, and handles failures in multi-step AI workflows.
01 ——
In plain English
Agent orchestration is how you wire multiple agents (or one agent with multiple tools) into a working system. Where a single LLM call answers a question, orchestration handles the messier reality: routing tasks to the right agent, waiting on long-running tools, retrying failures, and stitching results back together.
What orchestrators do:
- Route — send a sub-task to the agent best suited for it
- Manage state — track what's been done, what's pending, what failed
- Coordinate — let agents hand off work or work in parallel
- Recover — retry failed steps, fall back to a simpler agent
Common frameworks: LangGraph (graph-based agent flows), CrewAI (role-based teams), Temporal (durable workflows), Anthropic's Claude Agent SDK, and OpenAI's Swarm. Orchestration is where most production agent complexity lives — and where most failures happen.
02 ——
Related terms
AI Agent
Software that uses an LLM to plan and act — picking tools, taking actions, and adapting based on results to complete a user’s goal.
Agent Harness
The scaffolding around an LLM — tools, memory, loops, and orchestration — that turns a model into an agent.
Multi-Agent System
An AI architecture where multiple agents — often with different roles, models, or tools — collaborate on a task one agent could not handle alone.
Agent Swarm
A group of AI agents that work together — often with different roles — to solve a problem one agent could not handle alone.
Agentic Workflow
A multi-step task where an AI agent autonomously decides the steps, uses tools as needed, and works toward a goal with minimal human steering.
Workflow Automation
AI-powered tools that chain together multiple steps — apps, APIs, and AI models — to automate end-to-end business processes.
Back to glossaryLast reviewed May 2026