AI: Why I'm All In

I'm an AI maximalist. Not in the "slap a chatbot on everything" sense — in the "this fundamentally changes how we build and operate software" sense. I've spent the last 6 months really leaning into using AI as my primary working style and it's really taken off while building Swamp. Swamp is an adaptive automation platform for AI Agents. It's a CLI that supercharges those agents to create automation systems that are reviewable, shareable, and accurate. It may be built for agents, but it's there to empower humans. The work has made one thing undeniable to me: AI isn't a tool you bolt onto your existing stack. It's the new foundation you build on top of.

Ops Is Behind the Curve (Again)

Developers have embraced AI at speed — AI-assisted coding, AI-generated tests, agentic workflows for shipping features. Meanwhile, operations is doing what it always does: waiting. Watching from the sidelines, playing it safe and complaining about why it's not safe to use.

This is a pattern I've seen repeat across my entire career in infrastructure. Every major shift — configuration management, containers, infrastructure as code — these were all ops tools, built for ops problems. But every single time, it took the operations community years to fully embrace them. We build the tools and then drag our feet adopting the mindset shift that comes with them.

Ops is always late to the party, and AI is no different. But this time, the stakes are higher.

The speed at which AI is accelerating development is staggering. Code is being generated faster than ever. Infrastructure needs to support systems that are being developed at a pace that would have been unthinkable two years ago, never mind now. And all those hard-won best practices we spent years embedding into our organisations — security posture, cost management, compliance, observability, supply chain integrity — are being quietly eroded because the operational disciplines can't keep up with the velocity.

If ops doesn't get in on this early, it's going to be a shitshow. We'll end up with a generation of AI-generated infrastructure that's fast to deploy and impossible to secure, impossible to audit, and ruinously expensive to run. Everything we fought to fix over the last decade will unravel because we sat on the sidelines while the pace of change left us behind.

My passion is making sure that doesn't happen — closing the gap between what's possible and what the operations community is actually doing.

Why This Blog Exists

I started this blog because most of the conversation around AI is either breathless hype or defensive skepticism. Neither is useful. What's needed is practitioners showing their work — sharing what's actually possible when you commit fully to the idea that AI agents belong at the centre of software systems.

That's what I plan to do here. I'll write about what we're building with Swamp, lessons from the talks I've been giving, and the broader question of what software looks like when agents are first-class participants rather than afterthoughts.

If you care about infrastructure, about DevOps, about building systems that are reliable and maintainable at scale — this is the most exciting moment in the ecosystem in a decade. Ops doesn't have to be late this time. I'm not hedging, I'm all in. Let's fucking go.