Hacker News Reader

> He's advocating for manipulating the candidate into accepting the job by asking successively harder questions, even though you've already made up your mind. What a great start to your business relationship. And then justifies this by saying that 'people want to suffer'. This is almost like negging in romantic relationships.

And hazing.

Some form of the hazing seems very common for dotcom hiring right now, whether or not (as the article alludes at the start) the people who are mimicking it from elsewhere known the reasons or effects.

>- Personality is way more important than he makes it out to be. It can't just be broken down into 3-5 factors. This isn't a scientific process.

Eh, some would argue the opposite... (Probably hundreds or even thousands of psychologists & academic professors of psychology)

"In psychological trait theory, the Big Five personality traits, also known as the five-factor model (FFM) and the OCEAN model, is a suggested taxonomy, or grouping, for personality traits,[1] developed from the 1980s onwards."

https://en.wikipedia.org/wiki/Big_Five_personality_traits

"Abstract We demonstrate that Big-Five personality traits are stable for working-age adults over a four-year period."

https://www.sciencedirect.com/science/article/abs/pii/S01651...

"The five-dimensional structure was robust across major regions of the world. Trait levels were related in predictable ways to self-esteem, sociosexuality, and national personality profiles."

https://journals.sagepub.com/doi/abs/10.1177/002202210629729...

"Personality was related to career success controlling for general mental ability and, though adulthood measures of the Big Five traits were more strongly related to career success than were childhood measures, both contributed unique variance in explaining career success."

https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1744-6570....

There's intense critique of the Big Five model within psychology, and with what we've learned about the reproducibility of psychology studies , I'd be shocked if those studies you linked were reproducible.

I will ask successively harder and harder questions in an interview, time permitting.

However I do it only so that I can potentially give a hire recommendation for a higher level with tens+ thousands/yr more compensation.

Also because I'm always just one of 6+ interviewers, and a committee of 6+ completely separate people makes the actual decision, I can't even know if the person is being hired anyways.

The part about a candidate with multiple offers rejecting ones that felt “too easy” is not entirely unfounded. One of the things I’m personally interested in is professional growth so similar thoughts have definitely crossed my mind as a candidate

> Intelligence is way too diverse to be broken down into a linear scale

The article is written with this in mind, that's why there's multiple tests (coding and written!) for both "talent" and "judgement."

Intelligence isn't really very diverse, when you need people who are s-m-a-r-t smart. That means you need smart people, not half-smart differently-abled people.

As someone who has worked with engineers all day every day for the past 15 years, one thing that I don't see "measured" is how well that engineer can work cross functionality. I consistently see great results when cross functional collaboration is high and poor to catastrophic results when engineers are empowered to do everything themselves or where the org structure maintains engineering as an island unto itself.

> one thing that I don't see "measured" is how well that engineer can work cross functionality.

Large tech companies will measure for this. From personal experience I know that Google, Facebook and Netflix all test for cross functional leadership at the senior+ level. By the time you're interviewing for senior or higher roles, only two of the six interviews are actually about coding at these companies. The rest are behavioral and system design, with specific questions used to calibrate how hands on you are and how much you're used to working with many teams effectively in a large org.

Are you surprised given how the space he's in trains him to put out material like this (and actually think like this)?

Of course I'm reading this!

>IQ is not a good measure for intelligence. Intelligence is way too diverse to be broken down into a linear scale, even when it comes to just engineering or 'talent', as he calls it here.

Oh, you have a better way to measure g?

g is what you get when you only look for a single, linear scale.

The exact opposite, actually, it was discovered as a concept when people tried looking for multiple scales.

Maybe measurement is not the right approach here in the first place. Who says that everything can (or should) be measured?