I got asked LeetCode questions for a dev-ops systems engineering job today • Hacker News Board

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#211

Post by notyourwork » Fri, Jun 14, 2019, 5:11 PM UTC

mondoshawan wrote: post #148

notyourwork wrote: post #139

Earlier quoted context omitted.

It’s really not - as someone whose done 100s of interviews for a brand name tech company I don’t see this as a problem. The goal being for candidate to indicate their lack of knowledge but work through it. As an engineer you will encounter problems you don’t know how to solve. Pretending you know and bullshitting is a red flag.

There are hundreds of ways to do this without intentionally tilting the candidate into failure later. As an interviewer you have absolute authority in the room -- don't abuse it and destroy the candidate.

I think you are misunderstanding my point by indicating that interviewer's shouldn't abuse their power.

To be specific, I do a lot of things up front in an interview to help the candidate. The most important is informing them that they should treat technical exercises as a team exercise where we are teammates working together. Furthermore, I declare that there is no penalty for my assistance during the interview. The idea is for us to work together to solve a problem, my interest is in how they think and work through the problem. Furthermore, I explicitly tell them for discussion on prior experience and technical facets of interview that they are welcome to indicate they do not have an answer to my question or are unsure what's next.

All of this combined creates a very pleasant experience which is an important part of my goal as an interviewer. I'm not trying to embarrass people, though I am trying to make a informed hiring decision which is a very expensive endeavor for my company. Therefore, beyond letting candidate enjoy themselves, my main focus is to ensure that I have sufficient confidence to hire or be willing to walk away due to lack of conclusive data.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#212

Post by Stronico » Fri, Jun 14, 2019, 5:11 PM UTC

mondoshawan wrote: post #138

Stronico wrote: post #127

Earlier quoted context omitted.

So far it's worked very well for me. Since I've started using it (admittedly my sample size is small) I've found some very effective programmers, and automatically filtered out a lot of egotistical types. Software is usually a collaborative effort and this technique helps in that regard.

I've done well over 200+ interviews myself, and been on hiring committees where I have had chances to review other interviews. This kind of behavior usually trashes the following ones and aets the candidate up for failure.

As a recruiter or an employer? I've only done "employer type" interviews.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#213

Post by drugme » Fri, Jun 14, 2019, 5:15 PM UTC

What does this have to do at all with the job at hand?

Nothing whatsoever.

It really feels like there’s no connection to the reality of the role whatsoever anymore.

You got that right. Basically what's going on here is that the people running these interviews don't really know what they're doing -- and are secretly terrified of being "found out". So in a desperate attempt to compensate, they lunge for whatever technique they vaguely heard about being used at, you know, "leet" companies. And (as others have pointed out) at (apparently) very little cost or risk to themselves.[1]

Because that way... even though basically don't know why they're asking that question of you... and not only that, most likely would not be able to answer an analogously difficult (and unrelated to their own problem domain) question themselves! --

if you do manage to make it through, they'll at least know that you're "leet".

[1] Except of course, the opportunity cost of false negatives, and the reputation risk (and damage to recruitment prospects) that inevitably ensues once your company becomes widely known for mindlessly cargo-culting outdated and discredited interview techniques. Which they would know about, if they knew that they were doing. But then again, apparently they don't.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#214

Post by jimcsharp » Fri, Jun 14, 2019, 5:16 PM UTC

pault wrote: post #205

caymanjim wrote: post #85

Interviewing people is hard. Most people suck at it. I've interviewed over 100 people in my life, and I've been interviewed about ten times. I know I still suck at it. The code test is one of the more annoying facets of interviews. On the hiring side, it's an efficient way to screen and vet candidates for basic skills. It's relatively low cost to the employer to have a recruiter issue the test. If you come up with yo…

> My resume speaks for itself. At the time of this writing, I am working with someone who has 10+ years of experience, including former positions as a lead developer. I was involved in the interview process and made the recommendation to hire, as they were able to answer a lot of domain-specific questions very well and had a good knowledge of the ecosystem. Fast-forward three months, and I am spending hours per day s…

At least you have a constant source of code for interviewees to fix.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#215

Post by HorizonXP » Fri, Jun 14, 2019, 5:17 PM UTC

pault wrote: post #205

caymanjim wrote: post #85

Interviewing people is hard. Most people suck at it. I've interviewed over 100 people in my life, and I've been interviewed about ten times. I know I still suck at it. The code test is one of the more annoying facets of interviews. On the hiring side, it's an efficient way to screen and vet candidates for basic skills. It's relatively low cost to the employer to have a recruiter issue the test. If you come up with yo…

> My resume speaks for itself. At the time of this writing, I am working with someone who has 10+ years of experience, including former positions as a lead developer. I was involved in the interview process and made the recommendation to hire, as they were able to answer a lot of domain-specific questions very well and had a good knowledge of the ecosystem. Fast-forward three months, and I am spending hours per day s…

Would a better interview test be to ask the candidate to review code rather than write it themselves? That is, give them a code sample, describe what it aims to do, and ask:

1) Does this code accomplish the goal? 2) How would you improve this code to make it more readable? More efficient?

Personally, I think I'd rather work with people that know how to review code and provide feedback/commentary, since that usually means they can write good code themselves. It should lead to a better team dynamic.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#216

Post by Stronico » Fri, Jun 14, 2019, 5:19 PM UTC

UweSchmidt wrote: post #168

Stronico wrote: post #121

Earlier quoted context omitted.

Saying "I don't know" is an accurate answer. The ability to ask meaningful questions is a big part of any programming job. And (in my opinion) harder to learn than any particular technology.

Maybe you could give the critical replies to your comment some consideration. There seem to be many interviewers who abuse their temporary power over candidates by freestyling a psychological excercise, and brag about it in threads like this. In a way forcing the other party to give up with an "I don't know" is humiliating, and graciously allowing this answer doesn't make things better. You'd put yourself above the o…

I thought I had given consideration. I do not regard "I don't know" as giving up. Giving up would be not finding the answer eventually. I certainly don't find myself "under" some other person if they know the answer to a particular technical question.

As for interviewing a luminary of open source what would be the difference? I imagine they (much like everyone else) don't know lots of things and are probably comfortable saying so. Their answers to how they would go about finding an answer would probably be wonderful.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#217

Post by drugme » Fri, Jun 14, 2019, 5:20 PM UTC

Stronico wrote: post #110

When interviewing people I usually push until I get the candidate to answer "I don't know" on SOMETHING, just to see if they can do it. Not everyone can, and a lot of people will happily come up with a technical word salad to hide their perfectly understandable ignorance on some obscure technical problem I just came up with. That way you get to hear how they ask technical questions. Asking questions is a valuable (an…

When interviewing people I usually push until I get the candidate to answer "I don't know" on SOMETHING,

"just because I can". To me this technique comes off as condescending and manipulative, and smells like a power play basically. Not to mention - the better candidates will pick up on it, and your company will eventually get a reputation it doesn't want to have.

And really there's no need for it. Time and time again, I find that just by having a normal technical conversation with someone, on virtually any topic that they claim to have expertise in -- they will show their strengths and weakness (and propensity for BS, if any) very, very quickly.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#218

Post by sokoloff » Fri, Jun 14, 2019, 5:20 PM UTC

pault wrote: post #205

caymanjim wrote: post #85

Interviewing people is hard. Most people suck at it. I've interviewed over 100 people in my life, and I've been interviewed about ten times. I know I still suck at it. The code test is one of the more annoying facets of interviews. On the hiring side, it's an efficient way to screen and vet candidates for basic skills. It's relatively low cost to the employer to have a recruiter issue the test. If you come up with yo…

> My resume speaks for itself. At the time of this writing, I am working with someone who has 10+ years of experience, including former positions as a lead developer. I was involved in the interview process and made the recommendation to hire, as they were able to answer a lot of domain-specific questions very well and had a good knowledge of the ecosystem. Fast-forward three months, and I am spending hours per day s…

Is this person doing (valuable) work that would even require hiring a replacement? If not, fire them. If so, lobby to secure a backfill and then fire them.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#219

Post by positivejam » Fri, Jun 14, 2019, 5:25 PM UTC

skrebbel wrote: post #115

Stronico wrote: post #110

When interviewing people I usually push until I get the candidate to answer "I don't know" on SOMETHING, just to see if they can do it. Not everyone can, and a lot of people will happily come up with a technical word salad to hide their perfectly understandable ignorance on some obscure technical problem I just came up with. That way you get to hear how they ask technical questions. Asking questions is a valuable (an…

That's a trick question. Trick questions are bad because they test interviewing skill and not programming skill. It's a trick question because you want the interviewee to answer something different (I don't know) than what you're asking them to answer. Confident applicants might feel up to saying they don't know, but nervous people who are very self conscious in an interview setting may respond completely differently…

> nervous people who are very self conscious in an interview setting may respond completely differently than they would if they were asked the same question on the job by the coffee machine.

This part is crucial. If an interviewee is pushed to the brink but doesn't say, "I don't know," that does not imply they will not say, "I don't know," once they've secured a position.

Re: I got asked LeetCode questions for a dev-ops systems engineering job today

#220

Post by BryantD » Fri, Jun 14, 2019, 5:25 PM UTC

diminoten wrote: post #190

52-6F-62 wrote: post #108

Earlier quoted context omitted.

Hm. I spoke with Amazon (AWS) recently (haven't interviewed yet), and they made the express point that if I were to accept an interview that I should brush up on my algos and HackerRank questions because they loved HackerRank and would surely use a series of those-style problems throughout the interview process. My last in-depth interview was a take-home/collaborative project (paid) for a smaller company here in Cana…

It's probably per-team, someone on here mentioned that's how Google works at least. Someone I know did an AWS interview round during business school, and they do this thing called "bar raising" (IIRC) that involves one of the interviewers literally doing something unpleasant to see how you react (e.g. picking their teeth in the camera, cutting you off when you speak). Very odd.

That is not what bar raising is intended to be. There's always a bar raiser in an Amazon interview loop who is not affiliated with the hiring team. They ask the same kinds of questions as the rest of the interviewers in the same way, but they're not influenced by how much the team needs to make the hire, which in theory prevents the team from lowering the bar just to get a warm body in.

I'm not saying your friend didn't get a bad interviewer. It's just an interesting demonstration of how risky it is to generalize from anecdotes.