AI has enabled me to work through bugs, blockers, and new features significantly faster. From a product and business perspective, the deliverables are going out the door and the business is growing.

The other side of that coin is that I do feel like my personal ability to hand-write code from scratch and solve raw DSA problems has atrophied a bit. It's hard to say if this is actually a bad thing, or if from now on the role of the dev is shifting permanently toward a hybrid product manager / systems architect role.

At the same time, I have friends who have never touched a line of code suddenly spinning up "apps" to solve their own problems or curiosities. Inevitably, they all end up having to ask me or other software engineers about next steps, architecture, and what this or that error actually means.

At the end of the day, you still need a mental model of the system you're trying to build in order to actually maintain it. Whether you're using an axe or a chainsaw, you have to know which tree to cut, so to speak. The real question for me is: how do new devs build that intuition today if the chainsaw does all the early work?