The Point Was Never the Code

There's a lot of commentary right now from software engineers — people in the work, or who've spent their careers in it — about how the practice of software engineering has fundamentally changed. Writing code by hand, the argument goes, is no longer a viable long-term approach to building a product.

I understand why that lands hard. It's easy to get aggravated at the transition, or fearful of it. But I think it's worth stepping back and asking why alternative ways of building software might actually be desirable.

The point of creating code was never the process of writing code. It was building products, and ultimately solving problems. That's an important distinction. Code was always a means to an end — we just spent so long with our hands on it that we started mistaking the means for the mission.

AI writing the code today doesn't mean there are no problems left to solve. It means the trick has moved: it's about finding and discovering where the time is best spent. What is the next problem shaped like?

Software developers are used to taking on one project at a time — building a single web page, a single microservice. Now one person can build dozens of those things, or build systems that produce them independently. The leverage changed.

And here's what hasn't changed: AI hasn't devalued a human's ability to understand programming and systems — how to connect things, how to build relationships between components and between people, how to turn all of that into a product. What it has changed is the position from which we produce them.

We were never hired to type. We were hired to solve problems. That part of the job is still wide open.

Subscribe to Leadership Redefined: Master Adaptation & Conscious Strategies

Don’t miss out on the latest issues. Sign up now to get access to the library of members-only issues.
jamie@example.com
Subscribe