GitHub Copilot and AI Coding Tools in Practice
Год издания: 2025
Автор: Wienholt N.
Издательство: Apress
ISBN: 979-8-8688-1784-7
Язык: Английский
Формат: PDF/epub
Качество: Издательский макет или текст (eBook)
Интерактивное оглавление: Да
Количество страниц: 342
Описание: Learn the current state of Generative AI coding tools like GitHub Copilot, what the underlying models mean, and how to use them across the full development life-cycle. Look ahead to the near future of AI-generated software and understand how software developers can stay relevant in the industry.
Many companies have predicted that human coders will soon be redundant due to AI-generated code, but there is a big gap between the expectations and what is actually happening on the ground. A closer look at the current state of the tools and research in this area will offer realism and guidance to developers worried regarding redundancy.
Close the gap between hype and practical applications by receiving context and clear technical information on usage, understanding, and deployment of these tools.
This book addresses the gaps that exist between the flashy demonstrations and effectively using Copilot across the full range of the software development life cycle and covers the big gap that currently exists between agile methodology and successful using of Copilot and coding agents.
Copilot can be used effectively in many job roles, from data scientist and DBAs through to devops and architecture, and this book provides detailed coverage of how each of these job functions can take advantage of AI to increase their productivity.
The book concludes with an examination on the employment impacts of AI on software engineers, how they can adapt and leverage new skills to combat the threat that AI is believed to present to their job function, and practical ways that engineers can stay relevant in the new AI age.
The range of languages in the code bases that the code generation models are trained on is limited. While there are a huge number of projects for front-end work that developers have shared publicly to GitHub (when combined, Type Script and javascript represent the top language group), other “boring” languages like SQL have much less training data available. Python, as a more glamorous language, represents 16% of the code pushed to GitHub in Q1 2024, while SQL
doesn’t make the top 50 list.
What You Will Learn:
- How to use coding and software AI tools
- How software AI tools work
vHow software AI tools fit in an industry context
- How to use AI tools across the SDLC – it's more than just faster coding
Примеры страниц (скриншоты)
Оглавление
About the Author xi
About the Technical Reviewer xiii
Introduction xv
Chapter 1: Current State of Play: The High-Level View 1
Chapter 2: Using an AI Coding Agent 7
Chapter 3: Large Language Models: Under the Hood 29
Chapter 4: Prompt Engineering with AI Coding Agents 49
Chapter 5: Customizing and Extending Copilot 95
Chapter 6: Security in the Time of Copilot 121
Chapter 7: Designing Applications with Copilot 139
Chapter 8: Infrastructure, DevOps, and Monitoring with Copilot and AI 165
Chapter 9: Databases and AI 189
Chapter 10: Copilot and Data Science 201
Chapter 11: Code Migrations and Refactoring 251
Chapter 12: Test Augmentation with AI 275
Chapter 13: Management Challenges Introducing AI 297
Chapter 14: Surviving As a Software Engineer 307
Chapter 15: Introducing and Integrating Copilot in an Organization 317
Index 329
