The big theme was simple: a lot of companies are still talking about AI, while a smaller group are actually using it day to day.

Leigh and his team fall firmly into the second camp. They’ve rolled it out across hundreds of engineers and are learning by doing, not theorising.

We spent a lot of time on agentic engineering and what it really means in practice. The aim isn’t to replace people, it’s to remove friction so they can focus on what actually matters. In their world that means creating better guest experiences, not fighting software.

Code was never the moat, problem solving was. AI accelerates the output, but the real value still sits in judgement, design and engineering craft. Without that, you just get faster ways to build fragile systems.

There’s also a clear shift happening in how teams work. Product people now have real skin in the game. They can build, test and shape ideas directly, not just write requirements. That’s powerful, but it also raises the bar on knowing what’s actually worth building.

And underneath all of this is something Leigh called context engineering. Instead of repeating the same instructions to AI, they’re codifying standards and best practice so every output is consistent, secure and scalable.

On hiring, curiosity matters more than anything. If you’re not asking questions and digging into problems, you’re not going to get much out of these tools anyway.