Patil A., Patil M. - Practical Agentic Coding [2026, PDF/EPUB, ENG] torrent

Книги и журналы » Компьютерная литература » Программирование (книги)

Practical Agentic Coding

Год издания: 2026
Автор: Patil A., Patil M.

Издательство: Apress
ISBN: 979-8-8688-2670-2
Язык: Английский

Формат: PDF/EPUB
Качество: Издательский макет или текст (eBook)
Интерактивное оглавление: Да
Количество страниц: 719

Описание: Software development teams worldwide are struggling with a paradox: AI tools promise increased productivity, but many implementations result in technical debt, security vulnerabilities, and workflow disruptions. The problem isn't AI capability, but the lack of proper orchestration and governance. Practical Agentic Coding solves this challenge by introducing a strategic, structured approach to autonomous AI development.
This book teaches leaders and experienced developers on how to transition from using AI as a passive assistant to leading agents as autonomous team members. You will journey through a complete evolution: from setting a foundational industry vision and evaluating the market value of tools like Amazon Kiro, Google Antigravity, and Claude Code, to architecting intelligent workflows where AI agents handle complex tasks while humans provide strategic direction.
By combining the best architectural practices with hands-on tutorials, security frameworks, and real-world case studies, this practical guide enables teams to harness AI's transformative power without compromising quality or control. The result: development teams that operate at unprecedented speed and scale while maintaining the craftsmanship that defines exceptional software.
Cursor is not simply a fork of VS Code; it is a reimagining of the Integrated Development Environment where the Large Language Model (LLM) is the primary user interface. To truly master Cursor, you must understand the architectural pillars that allow it to “predict” code rather than just generate it.
Like Kiro and Antigravity, Cursor keeps VS Code’s extension ecosystem and UI while adding deeply integrated agents that can plan changes, edit code, run terminals, and call external tools through MCP. Below, we break down the key features of this AI-first architecture. In this section, we’ll go over the main tools and features of Cursor by going through some scenarios.
Most AI assistants use the “Fill-in-the-Middle” (FIM) paradigm, which involves looking at text before and after your cursor to guess the missing link. Cursor’s Tab feature has a fundamentally different architecture based on predictive coding. Tab is Cursor’s custom autocomplete that predicts multi-line edits and “next actions” rather than just the next few tokens. Inline Edit (Cmd+K) allows you to select a region and describe a transformation using natural language.
You Will Learn To:
- Understand the core principles of Agentic Coding and how to use AI Agents not just as tools but valuable team members
- Design workflows where AI agents can handle tasks like coding, testing, documentation, and more with human oversight.
- Evaluate the unique value propositions of leading tools like Kiro, Antigravity, Cursor, GitHub Copilot, and Claude Code
- Implement guardrails, audit trails, and governance practices to ensure safe and reliable agent behavior.
- Learn how to balance autonomy and control within integrated human/AI teams.
- Recognize and prevent common pitfalls like bad code generation, security issues, and broken workflows caused by poorly managed agents.
[spoiler="Примеры страниц (скриншоты)"]

[/spoiler]
[spoiler="Оглавление"]
About the Authors xxvii
About the Technical Reviewers xxix
Acknowledgments xxxi
Foreword xxxv
Introduction xxxvii
Part I: Foundations of Agentic Coding 1
Chapter 1: What Is Agentic Coding? The Evolution Beyond AI Assistants 3
Chapter 2: Key Concepts: Agent Memory, Rules, and Governance Frameworks 47
Chapter 3: The Human + AI Developer Model: Orchestration Principles 95
Chapter 4: Recent Advances in AI and Autonomous Agents 131
Part II: Agentic Coding Tools and Platforms 169
Chapter 5: The Architecture of Intent: Mastering Kiro 171
Chapter 6: Intent-First Development: Mastering Google Antigravity and the Antigravity CLI 217
Chapter 7: Cursor and GitHub Copilot (Agent Mode): Mainstream AI Integration 257
Chapter 8: Constitutional Coding: Mastering Claude Code 313
Part III: Architecting Agentic Development Workflows 357
Chapter 9: Agent Memory and Context Models: Building Cognitive Architectures for Agentic Systems 359
Chapter 10: Orchestrating Multiple AI Agents: Team Coordination 415
Chapter 11: Integrating Agentic Coding with DevOps and Cloud Infrastructure 453
Chapter 12: Security, Privacy, and Ethical Considerations in AI Development 49
Part IV: Real-World Implementation and Future Directions 539
Chapter 13: Case Studies: Successful Agentic Coding Transformations 541
Chapter 14: Migration Strategies: From Traditional to Agentic Development 573
Chapter 15: Measuring Success: KPIs and ROI for AI-Augmented Teams 609
Chapter 16: The Road Ahead: Preparing for the Next Wave of AI Development 643
Index 669
[/spoiler]