All shitcoders?

Question

Software design

Good afternoon! The question is this: good software products with good code-level architecture (i.e. relations of classes, modules, and not architecture in general, like a monolith, microservice, etc.) and the same good implementation in practice is a myth or reality ? My experience as a developer is not so great yet, only a year of commercial development, during this year I participated in two projects. And they both had shit code. Accessing the database directly in the controllers, bypassing the models, repositories or service layer (although there was such an opportunity, there were even ready-made methods for this, but requests were still written directly in the models), high connectivity of classes, where sometimes you need to disassemble to understand how the method works 5-6 classes associated with this method. Sometimes my senior comrades simply could not answer my questions about the code structure, how it works, although they wrote this code and the commit with this section was a week ago, or even less. And the question would not have arisen as such if during the lunch break these same people had not discussed high matters, about the quality of the code, about patterns, development principles, etc.
What would be interesting to know, this is not a question, but actually probably a survey: do you think that you are good at designing and implementing your idea in code? How objective is your confidence if the answer is yes?
The question is not for the sake of a holivar, I just very often hear from people philosophical arguments about what the code should be, but in reality these comrades write such that at least stand at least fall, or maybe just because of a lack of experience (after all, only a year of combat development) Should I think that it's not their code that's bad, but that I'm just not smart enough to understand their code? In general, I'm confused, help me figure it out, please.

Answer 1

basically it is.
business doesn't give a shit about our models and architecture, it needs to make money even yesterday.
if you lick the architecture from the start, you will never take off. and when and if you take off, the glade will already be sown with others who don't give a shit about the beauty of the code.
this is an iterative process: first we stake the plot somehow (MVP), then we live like this or put it in order.

Answer 2

The main thing for a business is to get results in a short time and preferably for a smaller budget. And "beautiful code" is the realm of nerds.

Answer 3

1. Make money for the business
2. Show that you can make money for the business
3. On the wave of success, find time for refactoring
4. Turn the shit code into a little less shit code
5. Repeat N times
6. Profit!

Answer 4

Good intentions pave the road to ... govnokod.
Here it is necessary to quickly, too lazy to look, I will fill in a direct request.
Here in this class I will make this field public. The test failed, okay, I'll add a separate if for this case. And here some crap is written, okay, I'll write another method that does the same thing, but in my opinion. Oh, and in this lib there is a new feature. And here they complain that slowly, FSUs, we will cache the data ... In general, even a well-conceived project eventually acquires all of this. I have never seen it differently (I saw a crooked architecture from the beginning, I also saw a complete lack of architecture as such).