Hacker News Reader

I can’t speak for anybody who made you an offer, of course, but often these kinds of problems are just another way to witness first hand how the candidate goes about solving problems. Failing to solve the problem given a short window of time is indeed not relevant, but your problem solving process is _extremely_ relevant. I’ve spent a lot of time fixing crap because some engineer was engaging in “magical thinking” that the cost of not doing such screening is clear. Brain teasers can be hit or miss, so I don’t particularly like giving those to candidates (though as an interviewee I loved them), but you can get a great idea of someone’s appetite for doing the mental work of problem solving through a thoughtfully-designed code screening.

I covered this in another reply, but if you use this process to see how I solve problems, you will conclude that I don't. You will witness me doing nothing, because I can't do anything in that circumstance.

Edit: nevermind, you already wrote about this elsewhere. Leaving the below here for context.

I'm curious, is this more of a social block or is it specifically the work medium (e.g. a whiteboard) that gets you blocked? I personally would happily let you use a laptop with whatever tooling you're most comfortable with, for instance. IMO whiteboards are okay for communicating high level architecture/ideas but abysmal at writing complex solutions (and IMO whiteboards imply you ought to have memorized the thing because of how punishing it is to need to rearrange content).

This is the claim, of course, that solving the problem is not what matters.

However, in practice, it is what matters, There is a "minimum" threshold (I can attest to this because I've been on the other side having to conduct them) where you more or less, must finish at least with some viable answer - even if unrefined - or you won't be moving on, full stop

I mean, sure, if you boycott the question you probably won't move on. But I'm not sure I want to work with someone unwilling to be curious or participate in problem-solving together. To me this filter is a positive effect of the test. In the interviews I've given I'll even handhold an applicant through to the optimal solution if need be, because to be perfectly honest I'd much rather have a coworker who is enthusiastic about learning and shows the ability to collaborate than someone who can write code but refuses to do so in a social setting (i.e. an interview). I want someone that I can have an intelligent conversation with because _I need my ideas validated, too_.

> refuses to do so in a social setting (i.e. an interview).

Interviews are very different from working together collaboratively. They're very different from presenting work to or working with a client or stakeholder, too, and even very different from a sales presentation. The space of things that might come up is effectively unbound, how you're being judged is wildly uncertain, you are being judged, and you know almost nothing about the people you're "working with". As practiced in software, they're closer to being called in to give a thesis defense without knowing in advance which thesis you'll be defending—and also everyone in the room is a stranger, and also you have no clue which aspects of your performance are being judged or by what criteria, and even know for a fact that some of the people conducting these have completely opposite opinions about which behaviors are desirable and which are "red flags".

I don’t think based on what I’m reading here that this is a criticism of coding during interviews so much as it is a criticism of candidate-hostile atmospheres in general. I’d also point out that social interactions are predicated on judging each other, and that this is healthy. It’s also not all one-sided. As an interviewee I choose to approach interviews as a process of satisfying my own curiosity and the curiosity of the potential employer about our viability as partners. That process doesn’t have to be a negative experience, and I don’t think a code screen or brain teaser really contributes directly to whether it is or not.

> In those days Apple did interviews the way they did everything else: each group at the company did things their own way. I don't know whether that's true anymore;

It's still true.

It sounds like you're misevaluating your technical interview performance. You can feel like you're "bombing" an interview, while still performing above expectation. Some interviewers ask hard technical questions, don't expect most candidates to complete them within the allotted time, and give "hire" recommendations for some (but of course not all) that don't.

I think a good interview puts you in a position where you are struggling and out of your expertise - being able to demonstrate you have a solution for a bunch of stuff is less important than being able to demonstrate how you do when you don’t know and have to figure it out.

In the general case, maybe you're right. In my specific case, "struggling" and "out of your expertise" doesn't describe it. It doesn't matter whether your question deals with my expertise or not; if it's technical and it requires me to actually think about it, I will be unable to answer in an interview setting. My mind will be completely blank.

Now, that's just me. Maybe I'm completely unique. Maybe there aren't any other programmers like me, and maybe nobody has to care about my individual quirks.

I don't have to care about them, either, because there seem to be enough people in the world who want to hire me. The only attention I need to pay to it is avoiding technical interviews to the extent that I can.

As a rule, when interviewing, I want to get to the boundary of a candidate's knowledge quickly. Questions they ace provide not much information (once it is established that they ace them), questions where they're totally out of their depth neither.

A bit provides maximum information when the chances for 0/1 are fifty/fifty.

So, agree with the sentiment

> a good interview puts you in a position where you are struggling and out of your expertise

as long as you are still at the boundary of your expertise, and have a realistic chance to make some progress.

> Questions they ace provide not much information (once it is established that they ace them), questions where they're totally out of their depth neither.

Agreed. The aim, for me, is to watch someone (try to) solve a problem and communicate about it. Ideally I'd like the question to rely on some previously unknown concept to see them pick up a new idea and run with it.

One of the best interview questions I ever had started with the interviewer asking me a fairly routine technical question, to which I gave a standard and acceptable answer. He then said, "Okay but what if I took away your ability to do X, Y, and Z - how would that change things?" He did that a few more times in different variations to really see how my answers would change especially as the tools/processes I would be forced to use started to stray from what I knew well.

I can see why it sounds that way. Let me supply some context.

I cannot answer any question in conversation with a stranger unless the answer is something I happen to know off the top of my head, or the question is about what I can see or feel or remember in the moment. It's related to a general sort of cognitive disability I have that shows up in several other ways.

For example, if I'm driving and you engage me in conversation, we will get lost. Every time.

If I'm playing a multiplayer game with voice chat, and people are talking to me, I will lose.

There are just certain cognitive activities that I cannot combine successfully. The presence of a stranger in conversation forcibly occupies all of my attention. I am unable to think about anything other than interpreting the stranger's utterances and preparing my own. If I try to, for example, answer questions about FizzBuzz that require me to actually think about how things work, my mind goes completely blank.

"I'm trying to think, but nothing happens."

I presume that in each case where I received an offer, it was because the team had information from other channels that made me attractive. I know it for a fact in a couple of cases.

So in recent years I've mostly relied on those other channels, and just skipped the so-called technical interview. All it's going to tell you about me is that I can't do what you want me to in a technical interview.

Interesting. I have this exact same problem. I gave up applying for regular jobs eventually as it was so demoralizing. Only work contracts through contacts now. It pays the bills, so I can’t really complain I suppose

It's nice to know that I'm not the only one.

You're definitely not. I often hear people describe it as multitasking. My wife excels at multitasking but I perform better having my complete concentration on a specific task (the one exception being listening to music).

I can't even have the TV on for "background noise" like some people like to because you can guarantee it will completely take my focus away from whatever else I'm trying to concentrate on.

Yeah, no tv and no music for me. Too distracting. I’m a musician, too, which just makes it worse.

This is the best argument I’ve heard against whiteboard / leetcode interviews. A lot of people have trouble talking and coding at the same time, it doesn’t make them less smart.

I'm pretty sure choking is behind most of the "LOL I caught a faker" stories from interviewers. I think it's far more common than people who've somehow been employed multiple places and can "talk the talk" but can't actually handle a for-loop. Most of them are just choking, under a very particular kind of pressure that's pretty much only ever encountered in interviews[0] and certain academic situations.

Like, there's the trope of the kid being called to the blackboard and not being able to solve some trivial problem even though they're not an idiot—and it's based in reality. IDK why we assume that reaction must be rare among adults.

[0] In certain kinds of tech interviews, to be specific. Somehow most of the rest of the white-collar and professional world gets by just fine without these kinds of hazing rituals.

In most interviews, if you say something like "Im just going to code in silence and we can talk about this later" is going to be well accepted.

How about, "I’m just going to take this question home to my office and work on it while out of all contact on no specific schedule so that I am not involuntarily preoccupied by my perception of a stranger watching me?"

No? I thought not. That’s why I simply avoid those kinds of hiring processes.

Does anyone even do brain teasers anymore? I recall them being popular 5-10 years ago but when I did my last round of interviews I didn't get a single one (across full interview loops with 6 Silicon Valley companies)