Most people believe that the software is built by logic, and this is why most developers spend their careers obsessing over the code: Clean Architecture, Design patterns, Performance benchmarks and so on.
Sure, all those things mentioned do matter. But I have started to believe that something matters more than all of it.
Taste
Contrary to popular belief, software is built by selection. By deciding what to keep, what to reject, and what to refuse to ship. While logic tells you how to build something, Taste tells you whether it should exist at all.
You see, code is learneable. Any dedicated person can learn how to write clean, efficient code. There are courses, books, documentations and even entire platforms built around teaching the craft. The roadmap is well-lit and well-mapped. To be honest, AI can already do most of this.
Taste, on the other hand, does not have a course.
Taste is the very difference between a developer who ships something functional and one who ships something that feels right. Between someone who adds features until the client stops asking and someone who pushes back because they know the product is getting worse. Between a developer who looks at a finished interface and says "we are done" and the one who says "something is still wrong here" even when they cannot immediately point at what.
Ira Glass described this better than I could:
"All of us who do creative work, we get into it because we have good taste. But there is this gap. For the first couple years you make stuff, it is just not that good. It is trying to be good, it has ambition, but it is not quite there. But your taste, the thing that got you into the game, is still killer. And your taste is good enough that you can tell that what you are making is kind of a disappointment to you."
That gap is real. Every developer who actually cares about their craft has felt it. You look at something you built and you know something is off but you do not have the skill yet to fix it. You look at someone else's work and feel the distance between where you are and where you want to be. That distance is not failure. It is proof that your taste is ahead of your execution. And that is exactly the right place to be, because it means you are still being pulled forward.
Now here is where it gets interesting.
We are living through a real shift in what building software actually means. Andrej Karpathy coined the term for it: vibe coding.
Andrej Karpathy
@karpathy
There's a new kind of coding I call "vibe coding", where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It's possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good. Also I just talk to Composer with SuperWhisper so I barely even touch the keyboard. I ask for the dumbest things like "decrease the padding on the sidebar by half" because I'm too lazy to find it. I "Accept All" always, I don't read the diffs anymore. When I get error messages I just copy paste them in with no comment, usually that fixes it. The code grows beyond my usual comprehension, I'd have to really read through it for a while. Sometimes the LLMs can't fix a bug so I just work around it or ask for random changes until it goes away. It's not too bad for throwaway weekend projects, but still quite amusing. I'm building a project or webapp, but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.
The idea is simple. You stop writing code and start directing a proposal engine. You describe what you want, the model generates options, and you decide what survives.
In the old world, the bottleneck was purely technical. Could you write the code or not? In the new world, you can generate ten plausible implementations in thirty seconds. The bottleneck is no longer whether you can produce a solution. It is whether you can recognize the right one. That judgment, that ability to look at ten generated options and know which one is actually good and which ones are just plausible, is not something a model hands you. It is something you build over years of caring about the craft.
Paul Graham
@paulg
Prediction: In the AI age, taste will become even more important. When anyone can make anything, the big differentiator is what you choose to make. https://paulgraham.com/taste.html
This means vibe coding does not make taste less important. It makes taste the only thing that matters. The developers who will stand out in the coming years are not necessarily the ones who write the most impressive code from scratch. They are the ones with the best calibrated judgment. The ones whose standard is high enough that they refuse to ship something just because it technically works. The ones who can look at a generated solution, immediately sense what is wrong with it, and know how to push until it is right.
The developers I actually admire are rarely the ones with the most impressive technical resumes. They are the ones whose decisions feel inevitable in hindsight. Who know when to stop adding. Who can say no to a feature not because they cannot build it, but because they understand it would make the product worse. Who look at a technically correct solution and reject it because something about it feels lazy.
They are not doing more. They are selecting better.
Code gets you hired. Taste is what makes the work worth looking at.