I have always had a weakness for leather shoes and well tailored clothes.
Not in a loud or logo driven way, but in the quiet details. The weight of good leather. The way a shoe creases after a few wears. The balance between structure and softness in something made to last. There is a particular satisfaction that comes from owning things that were clearly made with care by someone who knew exactly what they were doing.
Recently, I bought a new pair of leather shoes. Nothing experimental. Nothing trendy. Just beautifully made. As I held them, what caught me was not the brand or the price, but the craftsmanship. The stitching. The finish. The restraint. It was obvious that this was not the result of a template being filled in. Someone had made deliberate choices about what mattered and what did not.
That moment stayed with me longer than I expected.
At the same time, I had been wrestling with a very different kind of craftsmanship at work. AI generated code. Lots of it. Code that technically worked, but felt heavy. Verbose. Over engineered. Code that solved the problem, but did not respect it.
Standing there with those shoes in my hands, the connection became hard to ignore.
We already understand this dynamic in the physical world. Handmade shoes, bespoke suits, and hand sewn garments continue to command a premium even when cheaper alternatives are everywhere. This is not nostalgia and it is not anti automation. Factories are remarkable. They deliver scale, consistency, and access. But factories optimise for volume. Craft optimises for judgment.
AI today behaves much more like a factory than a craftsman.
It is extraordinarily good at producing output at scale. Text. Code. Ideas. Structures. It does not get tired. It does not hesitate. But it also does not possess taste. It does not know when something is sufficient. It does not feel the weight of unnecessary complexity.
This is where I see teams getting into trouble.
I have walked into codebases where AI has clearly done most of the writing. Everything works, but the system feels bloated. Endless layers. Excessive abstraction. Functions that say too much and mean too little. The code reads like a transcript of every possible thought rather than a set of intentional decisions.
The problem is not that AI wrote the code. The problem is that no one edited it meticulously.
In craftsmanship, removal is as important as creation. A good tailor does not just add fabric. They cut it away. A good shoemaker knows when a stitch is unnecessary. The value comes from deciding what should not exist.
That is the responsibility of the software engineer.
AI should not be asked to decide what you want. It should be asked to help you execute what you already understand. When you know the shape of the system you are building, AI can accelerate you. When you do not, it will happily generate something generic, verbose, and forgettable.
The highest leverage engineers are not those who produce the most code. They are the ones who enforce taste. They shape systems through constraint. They protect clarity. They treat software less like raw output and more like a finished object that someone else must live with.
Just as bespoke shoes coexist with factory sneakers, carefully crafted software will continue to matter in a world full of generated output. Abundance does not remove the premium on quality. It makes it more visible.
Engineers who work with care and precision will always command a premium.