People suck at technical interviews (2014)

Post by lmm » Fri, Jun 10, 2016, 4:26 PM UTC

Spoom wrote: post #145

Waterluvian wrote: post #90

Earlier quoted context omitted.

I hear you. I'm terrified by the kinds of interviews that want programming trivia. In the last three years I've never had to code a binary search tree. But I should be able to show you I'm very comfortable with Python.

Devil's advocate: Asking highly academic questions allows employers to be ageist without explicitly stating it, since college grads are likely to be the most familiar with them. This lets them select for younger candidates with less experience and thus drive down developer salary. (Standard disclaimer: Some development positions actually require the academic knowledge mentioned. Most of the time, though, a good devel…

Post by marcosdumay » Fri, Jun 10, 2016, 4:28 PM UTC

biztos wrote: post #33

PuffinBlue wrote: post #26

There needs to be a bit of a shift in understanding what 'category' of worker a programmer 'is', both by companies and more importantly by (some) individuals themselves. Programming is both technical and artistic. It's a creative endeavour that relies on technical skill to complete. The best analogy to another profession would be to those in the 'technical arts', those like photography, joinery, painting, sculpture,…

I agree completely, and this resonates with me particularly because I'm one of the many programmers also serious about other creative endeavors (in my case, the visual arts). But it's worth considering that in larger companies, management is often terrified of the creative aspect of programming, because it messes with their primary goal of predictability. Thus they seek, many of them openly and explicitly, to make al…

Post by lmm » Fri, Jun 10, 2016, 4:31 PM UTC

_asummers wrote: post #98

singingfish wrote: post #93

Earlier quoted context omitted.

I've recently got into the interviewing for devs game. Now we're a perl fairly specialised shop, and we only hire experienced people so that makes things relatively easy (candidates need to demonstrate a depth of knowledge, and . The pool of people around is relatively shallow too. We don't actually need to ask people technically detailed questions about specific algorithms. We have a conversation. From that conversa…

One of my favorite questions is asking people their least favorite and favorite languages then asking people to give one of their favorite features in the least favorite language, and least favorite features in their favorite language. Shows that they've actually spent time thinking about their tools. You get a surprising amount of insight from it, depending on the answer.

Post by lmm » Fri, Jun 10, 2016, 4:36 PM UTC

Clubber wrote: post #163

skrebbel wrote: post #2

In all honesty, I'm not sure the author did a lot of interviewing. For example, > The famous fizzbuzz test simply asks "are you aware of the modulo operator?" Wait, what? No, it asks "can you write a for loop without breaking a sweat?". There's a rightfully vivid debate going on about the virtue of asking algorithmic questions in interviews, but fizzbuzz is hardly algorithmic. I'd wager that virtually all programmers…

I think what he is pointing out is that if you don't know the modulus operator, you will fail FizzBuzz. The modulus operator is rarely used and somewhat obscure. The only time I use the modulus operator is when I'm writing something to track what record / rule I'm processing and I only want to show every 10,000th record or so. I can't think of a time I've used it otherwise.

Post by squeaky-clean » Fri, Jun 10, 2016, 4:39 PM UTC

benhoyt wrote: post #112

dfabulich wrote: post #21

Earlier quoted context omitted.

>> The famous fizzbuzz test simply asks "are you aware of the modulo operator?" > Wait, what? No, it asks "can you write a for loop without breaking a sweat?". There are notorious cases where senior engineers with years of actual honest coding experience "fail" fizzbuzz. IMO, That happens in two ways: 1) Not knowing (forgotten?) what a modulus operator is, and trying to code one up from scratch with a for loop. This…

If you don't know about modulus, you don't have to re-implement it. For Fizzbuzz you're looping anyway, so you can just maintain two counters, one that counts to 3 and resets, and one that counts to 5 and resets. When one of the counters resets, you know the main loop counter is a multiple of 3 or 5. Like so: def fizzbuzz_no_modulo(): counter_three = 1 counter_five = 1 for i in range(1, 101): output = [] if counter_t…

    from itertools import cycle, zip_longest, islice
    
    fizz_list = ['', '', 'fizz']
    buzz_list = ['', '', '', '', 'buzz']
    output = [
        fizz + buzz for fizz, buzz
        in islice(zip_longest(cycle(fizz_list), cycle(buzz_list)), 0, 100)
    ]
    for i in range(len(output)):
        print(output[i] or i+1)