3 ms·
Every few weeks someone comes back with the one true way of interviewing, or the X things wrong with how interviews are led. I have conducted a few hundred of t
by charles_f 5y ago
Every few weeks someone comes back with the one true way of interviewing, or the X things wrong with how interviews are led. I have conducted a few hundred of these by now, and the most I know about it is that there's no good way, because you try to figure someone out in just a few hours based on stuff they tell you. The format that seems to work the least worse for me is when you get them to tell you about actual stuff from their experience, which tends to prevent getting completely pointless people. I have been forced to do coding exercises for a while but I replaced by a general discussion on some tech they have been using recently, just to get a feeling of whether they understand what they're doing.
Recently I have been looking for another team internally to my company. An interesting fit is that I went through 3 interviews. I'm an engineering manager. Two interviews were focused on system design, one was coding. The only non coding question I received was around how I coach people. The three persons who interviewed me I asked: "what does the team need to do better", and they all answered a variation of "it takes a while to get stuff to prod once it's built. We need someone who can help get better at that". Yet not a single question for that. I guess the lesson learnt is that if you are looking for a particular skilk, maybe focus on that as well.
- wiseowise 5y agoThe is heavily biased to people who did X.
- charles_f 5y agoHow so?
- RNCTX 5y ago> they all answered a variation of "it takes a while to get stuff to prod once it's built. We need someone who can help get better at that". Yet not a single question for that. I guess the lesson learnt is that if you are looking for a particular skilk, maybe focus on that as well. If they knew enough about the problem and its solutions for it to filter down to people doing interviews, they wouldn't need to hire someone to teach them how to do it, right?
- charles_f 5y agoYou don't have to be an expert to get an idea of whether the person in front of you is experienced on a topic ; as long as you remain in the realm of their experience rather than going philosophic/theoritical - Can you tell me about a time when you had an issue delivering things to production quickly? What did you do to optimize? How did you know there was an issue? What alternatives did you consider? - Can you tell me about what you have done in the past to optimize they lead time to production? What strategies did you employ? How did you While you won't be able to tell if what they did was optimal, that can at least tell you if the person in front of you has some concrete experience in the topic, and some depth in their understanding. While that's not perfect, it's certainly better than flying blind