Nariman Mani, P.Eng., PhD Computer and Software Engineering

Software Testing with Generative AI

Engineering Reliable Quality Systems in the Age of Probabilistic Software

Software testing is entering a new era.

Generative AI is no longer a futuristic experiment—it is already shaping how test cases are created, how risks are analyzed, how failures are diagnosed, and how quality decisions are made. Yet most existing guidance treats generative AI as a productivity shortcut or a tool to "speed things up," without addressing the deeper engineering implications.

This book takes a different approach.

Testing with Generative AI is a rigorous, engineering-focused exploration of how generative models fundamentally change the nature of software testing—and what professionals must do to use them responsibly, safely, and effectively.

Rather than focusing on tools, certifications, or surface-level techniques, this book treats generative AI as a probabilistic software system that must be architected, constrained, validated, and governed with the same discipline applied to production systems.

Probabilistic systems demand stronger engineering discipline, not weaker standards.

Continuous Test Automation Engineering

Designing, Operating, and Evolving Test Automation in Modern Delivery Pipelines

This book is about treating test automation as a continuous engineering discipline, not a one-time implementation.

It focuses on how you design, integrate, operate, and evolve test automation within fast-moving software delivery environments.

What this book covers

How you engineer test automation as a system, not a collection of scripts

How test automation fits into CI/CD pipelines as a quality gate, not a bottleneck

How architecture decisions impact scalability, reliability, and maintainability

How to manage risk, flakiness, and technical debt in automated tests

How to use metrics and reporting to build trust with stakeholders

How to continuously improve automation as the system under test evolves

What makes this book different

It assumes you already know how to write automated tests

It focuses on engineering decisions, not tool tutorials

It addresses real-world constraints:

changing architectures

unstable environments

incomplete requirements

organizational pressure for speed

You learn how to adapt automation over time, without constantly rewriting it.

Who this book is for

Test Automation Engineers and SDETs

Test Architects and Quality Engineers

Senior Developers responsible for pipeline quality

Technical leaders designing delivery platforms

If you are responsible for keeping automation valuable month after month, this book is written for you.

Core idea

Test automation is never “done.”

Systems change. Pipelines change. Teams change.

Your automation must continuously evolve—by design, not by accident.

This book shows you how to engineer for that reality.