Recently, I stumbled upon the fast- filling post fillpackart , which prompted me to reflect, which I would like to share with all readers of Habr. However, first of all, I would like to emphasize that everything said in this article is my personal opinion and vision of the situation and I do not claim to be true.
The author of the above article complains that he was no longer asked technical questions at the interviews. Is it bad? How many of us write red-black trees every day at work or use virtual inheritance (these are just examples, please, let's not discuss their usefulness in the comments)?
It seems to me that there are not so many people who answered the last question positively. But it is precisely about this that they like to ask at technical interviews. Will a business suffer badly if it hires a person who cannot answer such a question without a search engine at hand?
The problem is that the questions on technical interviews are strikingly different from the real tasks of the developer . Including from the tasks that the candidate will perform in the company, if it passes the interview stage. So why this? Why check the ability to solve puzzles with LeetCode or knowledge of the device of exotic language constructs? Why is this needed, if in the future it is hardly useful, and if it is useful, you can always find a detailed description of these algorithms and principles?
And most importantly: it is not just a useless check. It turns the interview into a meaningless and merciless hell. Because of this, the preparation for the interview turns into, in fact, a useless solution of the tasks. At the interviews themselves, the nerves take their toll and even the most worthy candidates are sometimes lost in solving simple tasks.
Here, it must be said to me that the purpose of the interview with the tasks is, first of all, to check the candidate’s thinking . Yes, it is, but not everyone understands this. Many developers, whom HR sends to interview applicants, forget about it. And then the interview turns into a boring monologue of the candidate who is trying to solve something. Then the decision is photographed and it becomes an argument for or against. Is it correct? Not. I believe that the tasks can be given only simple ones and the candidate must be helped. It is not necessary to tell a person the whole decision, but you need to push him to the right thoughts. After all, we are not developing software alone. Development has long been a team work. So why do I still see developers who do not say anything during the interview except the condition of the problem?
Once again I would like to focus attention: interviewing puzzles is normal, but only if these puzzles are simple and its solution turns into a dialogue between the interviewee and the interviewee. The main thing here is to check how the candidate thinks. How he approaches the problem. That's what's important, and not whether he solves this problem or not. You are not looking for developers whose only skill is solving problems with LeetCode?
Near there is another question. Is it true that it is better to hire a good introvert than a mediocre extrovert (I’m not talking about hiring mediocre employees, but you can see such people during an interview)? I'm sure not: as I said before, the development has long been a command, and then the interaction of people in a team becomes important. Is it really better for your partner to sit for a week and understand some class or function from the general code base than if he approached his developer and asked him how to use it correctly? Here we can say that the cool introvert is also not clear how to respond to questions to him.
Single development dies out. In the agony, only freelance site developers jerk. Any serious product requires a development team, and often not one. Therefore, just cool introverts will soon die out of IT, and we will have to work with what is left.
The only point in which I agree with the author of the article, but even here is not completely. Yes, there is a problem with mediocre and furious technologies. But even here it is important to note that it's better that everyone uses one thing. The only solution for sure sooner or later will be the most elaborate. If only because its developers have the money and resources for it. Otherwise, we will get many solutions that started well, but in the course of development we acquired various problems. This state of affairs only prompts the development of the "14th standard". And so on. Let's all still use one solution. Someday his problems will be solved, which cannot be said about each of the 10 different solutions with individual users.
Source: https://habr.com/ru/post/452372/
All Articles