Agentic Coding with LLMs: Build Autonomous Software Generative AI and Advanced Prompt Engineering

What happens when an AI system can do more than generate a function? It can inspect a repository, understand a development task, create a plan, modify multiple files, run commands, execute tests, analyze failures, and continue working until it reaches a solution. Agentic Coding with LLMs: Build Autonomous Software with Generative AI and Advanced Prompt Engineering explores the technologies and engineering principles behind this new generation of autonomous software development systems. This book takes readers beyond traditional AI coding assistants and into the architecture of coding agents. You will learn how LLM-powered agents reason about development tasks, break complex goals into smaller actions, use external tools, manage memory, retrieve relevant information, and evaluate their own results. A major focus is placed on advanced prompt engineering and context design. Readers will discover how system instructions, task decomposition, structured outputs, retrieval, feedback loops, and verification strategies can dramatically improve the reliability of autonomous coding systems. From single-agent workflows to collaborative multi-agent architectures, the book provides a practical framework for building AI systems capable of performing meaningful software engineering tasks. You will also explore testing, security, observability, failure recovery, and human oversight critical components for moving agentic coding from impressive demonstrations to dependable engineering tools. For developers, AI engineers, and builders preparing for the next era of software development, this book provides the knowledge needed to design, build, and control autonomous coding systems powered by Large Language Models