liveBook · Manning

Agentic development

Also called agentic coding

Agentic development is a software-development approach in which AI agents perform or coordinate programming and related engineering tasks, autonomously or with human oversight.

How 3 Manning books explain it

The Art of Code

A software-development approach in which AI agents perform or assist with programming tasks across one or more steps.

How it is used
Uses AI agents to accelerate programming while emphasizing human judgment in producing durable, high-quality code.
How deep it goes
Explains quality risks and their causes, but does not provide a step-by-step supervision procedure.
Examples
Contrasts clean and tangled surrounding code, repeated corrective patches, and working code that nobody fully understands.
What you can do after
Evaluate agent-generated code by recognizing quality mirroring, patch accumulation, and comprehension debt.

Read how The Art of Code explains agentic development →

Claude at Work

Agentic coding is a software development approach in which an AI agent directly navigates, modifies, builds, deploys, and tests a repository's codebase.

How it is used
Uses an AI agent inside a repository to navigate, modify, build, and test software.
How deep it goes
Defines repository access and multi-tool work, then addresses autonomy, permissions, task fit, and human review.
Examples
Demonstrates dotnet build, npm start, pytest, and investigating a Kubernetes deployment against YAML.
What you can do after
Evaluate AI-generated changes by choosing task autonomy and reviewing build and test results.

Read how Claude at Work explains agentic development →

Introduction to Generative AI, Second Edition

Agentic coding is software development in which AI agents plan and execute multistep programming tasks with limited human intervention.

How it is used
Deploys AI agents within software development workflows to pursue coding goals across files, tests, and debugging tasks.
How deep it goes
Explains capabilities and workflow types at a high level, stopping short of implementation details or agent internals.
Examples
Uses bug fixing, API modernization, test generation, OAuth implementation, and database migration cleanup as delegated tasks.
What you can do after
Apply agentic workflows through Copilot, Cursor, or Windsurf to delegate, test, debug, and refine multistep code changes.

Read how Introduction to Generative AI, Second Edition explains agentic development →

Also covered in

What to know first

Related concepts

Questions about agentic development

What is agentic development?

Agentic development is a software-development approach in which AI agents perform or coordinate programming and related engineering tasks, autonomously or with human oversight.

Which Manning books explain agentic development?

The Art of Code, Claude at Work and Introduction to Generative AI, Second Edition.

What should I know before learning agentic development?

Manning books that teach agentic development usually introduce artificial intelligence and specification first.

What is the difference between agentic development and traditional software development lifecycle?

Agentic development is a software-development approach in which AI agents perform or coordinate programming and related engineering tasks, autonomously or with human oversight. A traditional software development lifecycle is a software process in which work proceeds through stages with substantial human documentation and implementation effort.

What is agentic development also called?

Agentic development is also called agentic coding.

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.