Home · Articles


Programming and creativity

November 21, 2026

When people say programming involves creativity, it is easy to misread that as comparing software to art. Programming is deeply technical, measurable, and accountable to reality. Code either compiles, scales, recovers from failure, and holds up in production, or it does not.

Still, a large part of being a good programmer is creative work. It is creativity in the sense of technical craft.

You see it most clearly in the places where there is no single “correct” answer. Give two experienced engineers the same problem and the same requirements, and they may come back with completely different designs. Not just different names, but different boundaries, different data models, and different assumptions about what will change later. One design makes the next three years of work feel natural. The other turns every small change into a fight. Getting to the first one is rarely a purely deductive step. It takes imagination to see a cleaner shape before it exists.

Getting to a decent implementation is rarely a purely mechanical derivation. You cannot apply a checklist and expect the right abstraction to fall out. A lot of it is being able to see a few steps into the future, sensing where the system will bend, which assumptions will age poorly, and which boundaries will still make sense two years from now. That relies on intuition as much as analysis, and it draws on personal creativity just as much as experience. Two people can carry the same years of experience and still see completely different paths forward, because the leap to a simpler shape is personal.

That is where craft lives. It shows up in the details that are easy to overlook, like making invalid states unrepresentable, choosing a data structure that eliminates an entire synchronization problem, keeping an API small enough that callers cannot easily misuse it, or deleting a layer of indirection that stopped paying for its own complexity. None of those choices are decorative. Every one of them is a technical decision, and every one of them requires judgment.

Craft is also about restraint. Knowing how to build a complex system is a technical skill; knowing how to solve the same problem with half the moving parts is craft. You do not get that instinct from reading specs. You build it by writing code, running it in production, debugging your own mistakes, and paying attention to why certain designs stay solid under pressure while others become brittle.

As more of the routine code generation gets automated, this side of programming matters more, not less. Producing lines of code was never the hardest part of the job. Seeing the right decomposition, shaping the invariants, and exercising technical craft over what gets built is where the real work still sits.


Home · Articles