AI can already write code. Often faster than we can, and sometimes better. It makes many developers quietly wonder what this means for their careers. If the code is becoming the easy part, what exactly is our job now?

This talk comes from my own work wrestling with these questions as the owner of a software consultancy, trying to understand how to be ready to keep proving value to our customers in the age of commonditized code.

In this talk, we will look at what AI is genuinely good at in software development, where it consistently falls down, and how those strengths and weaknesses are already reshaping the role of developers. You will learn how to use AI in ways that make you and your teams more effective, and how misusing it can slowly erode judgment, ownership, and long-term code health.

We will also talk about how this shift changes career paths. That includes the challenges facing entry-level developers, how senior developers’ responsibilities are evolving, and what experienced teams are focusing on when code is no longer the bottleneck.

The goal is not reassurance and it is not panic. It's clarity. When the code writes itself, developers remain essential, but the value moves away from typing and toward understanding problems, making tradeoffs, and shaping systems that hold up in the real world.