Agent orchestration
Orchestration is the coordination and sequencing of agents, services, tasks, or components to carry out a larger workflow or objective.
Generative AI in Action
Container orchestration is the automated deployment, scheduling, scaling, networking, and management of containers.
- How it is used
- Deploys, manages, and scales containerized microservices that support enterprise generative AI applications.
- How deep it goes
- Explains service and workflow responsibilities plus operational concerns, but does not give a Kubernetes configuration or deployment procedure.
- Examples
- Uses Kubernetes as the container-management example and an enterprise content-creation pipeline to illustrate coordinated AI work.
- What you can do after
- Evaluate/understand where container orchestration supports a generative AI stack and where application-level workflow orchestration remains necessary.
Read how Generative AI in Action explains agent orchestration →
AI Agents in Action, Second Edition
An orchestrator is an agent that maintains a plan, delegates subtasks to worker agents, evaluates their results, and decides how execution continues.
- How it is used
- Coordinates specialized workers when one agent faces multiple goals or needs to choose and revise subtasks dynamically.
- How deep it goes
- Develops the pattern into a typed plan-and-decision loop with worker delegation, evaluation, replanning, and an iteration limit.
- Examples
- Uses research and analysis workers, a flow converted to agent-as-tool delegation, and a solar-versus-wind comparison exercise.
- What you can do after
- Build it from scratch as a bounded loop that delegates research and analysis, inspects decisions, and aggregates findings.
Read how AI Agents in Action, Second Edition explains agent orchestration →
AI Agents and Applications
A supervisor agent is an agent that coordinates a group of more granular agents to accomplish a larger task.
- How it is used
- Coordinates specialist agents when a multipart request requires results from more than one domain.
- How deep it goes
- Works through specialist setup, supervisor configuration, invocation, and traced handoffs; does not detail the supervisor’s internal decision algorithm.
- Examples
- Uses a Cornwall weather-and-hotel request, named travel and booking agents, a supervisor configuration, and an execution trace.
- What you can do after
- Apply it via LangGraph’s create_supervisor to coordinate named specialists and diagnose their handoffs with LangSmith.
Read how AI Agents and Applications explains agent orchestration →
Effective Platform Engineering
A container orchestration platform manages the deployment, scheduling, scaling, and operation of containerized workloads.
- How it is used
- Coordinates application operations and reduces manual infrastructure work while supporting platform consistency and reliability.
- How deep it goes
- Explains Kubernetes’ desired-health loop and cluster roles, then discusses platform orchestration without detailing its implementation.
- Examples
- Shows Kubernetes control planes and worker nodes, plus Humanitec enabling self-service application deployment.
- What you can do after
- Understand how orchestration maintains application health and abstracts infrastructure tasks from developers.
Read how Effective Platform Engineering explains agent orchestration →
Kubernetes for Developers
A container orchestrator is a platform that schedules, operates, scales, and maintains containerized workloads across compute resources.
- How it is used
- Uses Kubernetes to schedule and monitor containers across shared compute rather than managing each machine separately.
- How deep it goes
- Defines scheduling and monitoring, then explains deployment attributes and operational benefits without detailing implementation internals.
- Examples
- Illustrates multi-zone services, service rollouts, and deployments ranging from databases to batch jobs.
- What you can do after
- Operate deployments by specifying replica counts, availability needs, and service networking instead of choosing machines.
Read how Kubernetes for Developers explains agent orchestration →
Acing the CCNA Exam: Volume 2 Advanced Networking and Security
A container orchestrator is software that automates the deployment, management, and scaling of containers.
- How it is used
- Automates deployment, management, and scaling when applications use many containers or independently managed microservices.
- How deep it goes
- Defines the operational purpose and deployment scale, but does not detail Kubernetes architecture, commands, configuration, or troubleshooting.
- Examples
- Uses Kubernetes as the industry-standard example and Docker Engine as the container-engine example.
- What you can do after
- Evaluate when automated container deployment, management, and scaling is preferable to manual operation.
Introduction to Generative AI, Second Edition
An orchestration framework is a software framework for coordinating agents, tools, memory, roles, and task execution.
- How it is used
- Coordinates agent roles, tool access, memory, and task execution within dynamic systems.
- How deep it goes
- Outlines configurable components and multi-agent coordination, but stops short of implementation details.
- Examples
- Names AutoGen, LangChain, LangGraph, the OpenAI Assistants API, Swarm, the OpenAI Agents SDK, and Azure AI Foundry.
- What you can do after
- Understand how orchestration coordinates specialized agents and why autonomy raises control and evaluation challenges.
Read how Introduction to Generative AI, Second Edition explains agent orchestration →
Data Pipelines with Apache Airflow, Second Edition
A container orchestration system is software that coordinates the deployment, execution, scaling, and distribution of containers across multiple machines.
- How it is used
- Coordinates container deployment and scaling across machines, including running Airflow tasks in Kubernetes pods.
- How deep it goes
- Explains Kubernetes components, scheduling considerations, and pods at a high level, but stops short of a full Kubernetes account.
- Examples
- Uses Airflow tasks running in Kubernetes pods and contrasts cluster execution with Docker on a single machine.
- What you can do after
- Apply Kubernetes to run containerized Airflow tasks across worker nodes.
Read how Data Pipelines with Apache Airflow, Second Edition explains agent orchestration →
Podman in Action: Secure, rootless containers for Kubernetes, microservices, and more
A container orchestrator is a system that schedules and manages container workloads across one or more machines.
- How it is used
- Coordinates container workloads across multiple machines by directing container engines to run them on selected nodes.
- How deep it goes
- Distinguishes orchestrators, engines, and runtimes, but stops short of deployment procedures or scheduling internals.
- Examples
- Names Kubernetes, Docker Swarm, and Apache Mesos as orchestrators, with CRI-O and containerd serving Kubernetes.
- What you can do after
- Evaluate whether a workload needs a single-host container engine or a multi-machine orchestrator.
Financial AI in Practice
Orchestration is the coordination and scheduling of multiple tasks, processes, or services in a workflow.
- How it is used
- Coordinates tool selection in financial AI systems and dependent stages in production quantitative-data pipelines.
- How deep it goes
- Explains tool sequencing, dependency management, retries, scaling, and observability without specifying an orchestration implementation.
- Examples
- Uses financial question answering with retrieval and live API calls, plus market-data pipelines running from ingestion through reporting.
- What you can do after
- Operate dependent pipelines by coordinating stages, retrying failed jobs, scaling resources, and making execution observable.
Read how Financial AI in Practice explains agent orchestration →
Building Reliable AI Systems
The coordination and sequencing of multiple task steps to accomplish a larger objective.
- How it is used
- Coordinates tools and task steps so agents can carry out complex requests rather than only respond to them.
- How deep it goes
- Explains task sequencing, ReAct cycles, and framework trade-offs without detailing orchestration code.
- Examples
- Uses travel planning, stock-price retrieval, and meeting scheduling to illustrate coordinated agent tasks.
- What you can do after
- Apply LangGraph to controlled branching or cyclic workflows, or CrewAI to straightforward linear sequences.
Read how Building Reliable AI Systems explains agent orchestration →
Related concepts
- Kubernetesincludes · Kubernetes is an open-source platform for orchestrating and automating the deployment, scaling, and management of containerized workloads across computing clusters.
- containeruses · A software container is an isolated execution environment that packages an application and its dependencies while sharing the host operating system kernel.
- Docker Composeincludes · Docker Compose is a tool for defining and running applications composed of multiple containers through declarative configuration.
- distributed systemkind of · A distributed system is a computing system whose components run across multiple networked processes or machines and coordinate through communication.
- AI agentuses · An AI agent is a computational system that uses artificial intelligence to interpret information, select actions, and pursue goals through interaction with an environment.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.