There have always been two kinds of developers. Some of us fell in love with the code itself: the elegant class hierarchy, the clever one-liner, the satisfaction of a perfectly organized solution. Others mostly wanted to solve interesting problems for people, and software just happened to be the best tool available. There's nothing wrong with either kind, and most of us are a mix of both. But AI is only coming for one of those job descriptions.

The answer to the title question, by the way, is yes. The employment data doesn't show developer jobs disappearing. It shows the job changing shape underneath us. Producing syntax is getting cheap, but what stays stubbornly human is the judgment: knowing what to build, how to architect it, which tools and platforms actually fit the constraints, and how the thing should work, so you don't end up replacing a bad process with a faster, automated version of the same bad process. Those calls were always the hard part of this job, but now they're becoming most of it.

I've spent nearly three decades building software and the last ten hiring the people who build it, and what I look for in a developer has changed more in the last two years than in the twenty before.

In this talk we'll look at what the hiring data actually shows, which developer skills are in gaining value and which are becoming commoditized, and how to grow toward the problem-solving end of the spectrum while there's still plenty of runway.