Introduction

Why This Book?

Every organisation is being told they need AI. Vendors promise transformation. Headlines predict disruption. The pressure to act is immense.

But most AI initiatives fail — not because the technology doesn't work, but because organisations don't know where to start, what to build, or how to govern what they've built.

Agentic AI — artificial intelligence that can pursue goals autonomously, make decisions, and take actions — represents the next frontier. It's genuinely transformative. It's also genuinely risky if approached without discipline.

This book exists to cut through the hype and provide a pragmatic path forward.

Who This Book Is For

This book is for business leaders, technology professionals, and practitioners who need to understand agentic AI — not as a theoretical concept, but as a practical capability to be designed, built, and governed.

You might be:

Whatever your role, this book will give you the foundations, patterns, and practical guidance you need.

The Pragmatix Perspective

This book was developed by Pragmatix, a Brisbane-based digital transformation consultancy. We've led transformation programs worth over $300 million across healthcare, government, financial services, and not-for-profit sectors.

We've learned — sometimes the hard way — what works and what doesn't when it comes to aligning technology with business strategy.

Three principles guide everything we do:

1. Business intent drives technology decisions — not the reverse.

Technology is an enabler, not the goal. Every investment in agentic AI should trace directly back to a business outcome. If you can't articulate the business value, don't build it.

2. Game of Inches — incremental value over big-bang transformation.

Sustainable change happens through consistent, small wins — not risky, all-or-nothing programs. Start small, prove value, then scale. This is how you build momentum while managing risk.

3. Re-use before buy, buy before build — especially with AI.

Your existing platforms are racing to embed AI where your data already lives. Salesforce has Einstein, ServiceNow has Now Assist, Microsoft 365 has Copilot — these vendors know the data model and are investing billions. Use that AI first. Don't extract data into a central hub just to build what your vendors are already building. Buy proven AI solutions where gaps remain. Build custom only when you have a genuine differentiator that no one else can provide.

4. Stay curious — innovation is no longer hostage to complexity.

For decades, organisations have been held captive by software vendors who relied on complexity to lock in customers. Those days are ending. Agentic AI teams can now create custom software tailored to your specific needs faster than ever before. But remember: software built must be software supported. You can't do it all yourself, nor should you. Find strategic partners who help you innovate at scale — without the hefty price tag of traditional system integrators. The right partner accelerates your capability, not your dependency.

The Pragmatix Digital Transformation Framework

These principles are embedded in the Pragmatix Digital Transformation Framework (pX-DTF), our methodology for guiding organisations from strategy through to sustained value realisation. This book applies that same pragmatic thinking to agentic AI.

How This Book Is Structured

The book is organised into four parts:

Part 1: Foundations
What agentic AI is, how it differs from traditional AI, and the building blocks you need to understand. We introduce the Pragmatix Agentic AI Maturity Model and explain the technology stack — what you rent, what you buy, and what you build.

Part 2: Patterns
Reusable architectural patterns for agentic systems — how single agents reason and act, how multiple agents work together, how humans stay in the loop, and how agents access and retain information.

Part 3: Production
What it takes to move from prototype to production — evaluation and testing, security and guardrails, cost management, and ongoing operations. The practical realities of running agents in the real world.

Part 4: Transformation
The business case for agentic AI, how to identify automation candidates, governance for autonomous systems, and a roadmap for your transformation journey. This is where technology meets strategy.

How to Use This Book

You don't have to read this book cover to cover.

Wherever you start, remember: this is practical guidance, not academic theory. Use what's useful. Adapt it to your context. And if you want help applying it, that's what Pragmatix is here for.

Let's begin.

☰ Contents