Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

How to Hire

blog.samaltman.com

Post reply on HN
Print view View original

41–50 of 82 posts

Re: How to Hire

#41

Post by mkramlich »

Scorponok wrote: post #13
CraigJPerry wrote: post #9

I'd love if more companies adopted auditioning over the traditional interview process. It was counter intuitive to me at first but it's potentially a cheaper way to hire. There's less downtime for the hiring team - everyone can more or less continue what they're working on. It's a better quality of interview too - candidates have a more realistic platform to show their worth.

For candidates who already have jobs, how would "auditioning" work? Is it assumed that they'd take time off from their current job to do it?

An audition can be a small paid contract that you can work on in your otherwise free time hours, here and there, on your own premises, then just do whatever interaction or delivery is necessary to show results and demonstrate value. And you get paid for it. It doesn't have to be a MF/office-day/on-site thing. The goal is just to test out two-way fit under as realistic of circumstances as possible, testing for the things you care about most, and, really, have it be reciprocal: the applicant/candidate gets paid. The details are choosable by both parties and should be mutually agreeable. Sprinkle to taste. And you can iterate.

Re: How to Hire

#42

Post by ohazi »

farmdawgnation wrote: post #32
ohazi wrote: post #22

> Whenever possible (and it’s almost always possible), have someone do a day or two of work with you before you hire her; you can do this at night or on the weekends. No. Absolutely not. If a candidate really is as good as you think he/she is, they already likely has a full time job that they're perfectly happy with (as you already well know): > Often, to get great people, you have to poach. They’re never looking for…

Sorry, but speaking as a software engineer who does have a bit of talent, I disagree with you in saying that this type of audition exercise is below me. We complain and complain and complain as an industry about how the interview process that The Big Boys still use (which is essentially just a pop quiz of vocabulary terms at its core) is broken. Why is this not a better alternative? You wouldn't pitch this as "hey, I…

> I disagree with you in saying that this type of audition exercise is below me. > We complain [...] is broken. Why is this not a better alternative?

It's not a better alternative because many of the best developers in Silicon Valley don't disagree with me. There are many reasons for this -- not everyone is a carefree recent college grad, and many people can't afford the luxury of taking the time off to do something like this. You might argue that it's a paid trial, but often "can't afford" has more to do with family obligations, time commitments, and mental bandwidth than money, even if they'd be more than willing/able to push hard in a start-up environment if it were their full-time job. Many of the best developers aren't willing to leave their current job or take time off without the commitment of a written offer. They may be more risk averse than someone younger with fewer obligations, for many of the same reasons outlined above.

But the most important reason why this is not a better alternative is that any developer who doesn't want to put up with this doesn't have to put up with it, since it's by far not the norm. Any developer who feels this way and is actually any good would have an easy time of finding an equal or better opportunity elsewhere.

> You wouldn't pitch this as "hey, I'm still not sure about you so I'm going to ask you to do X to prove your worth" - though it seems that's the interpretation you're getting from this.

It really doesn't matter how the company tries to spin this. In the current environment, the best developers in Silicon Valley can work wherever they want. By using a paid trial as policy, these companies are (willfully or not) turning away some of the best developers that they interview, for very little gain.

tl;dr: It doesn't matter whether or not you perceive this exercise as being below you. If a large enough contingent of the best developers won't put up with it, the companies who use the exercise are losing out.

Re: How to Hire

#43

Post by larrys »

ig1 wrote: post #37
larrys wrote: post #30

Earlier quoted context omitted.

"The absolute cheapest benefit you can offer an employee is their title, and if they have a rational reason for wanting a better title, why on earth would you refuse to give up some title to get a good candidate on board?" Well one reason [1] might be that by giving the a "big" title it will enhance their resume in the eyes of another company that is looking to poach. Doesn't matter if it works out or not at the new…

Anytime your reasoning is "we shouldn't do it because it will make them attractive to other companies" you're probably wrong. You want a company full of super-stars that everyone wants to poach. Otherwise you're just like the companies that prohibit employees from have linkedin profiles or don't give employees training because that makes them more poachable. Crazy.

Not sure you understand what my point is.

I'm not saying you shouldn't have employees that are so good others want to poach them.

I'm saying that by giving someone a title (that they might not deserve) you are possibly making them more attractive and more likely to be poached. This was said simply to counter the parent point that "there is no cost".

Let's say I allow you to put on your card "VP of Software Engineering" vs. just "Software Engineer" as only one example. (Forget legalities I am using this to prove a point).

Or, "National Marketing Manager" instead of "Marketing Manager".

Re: How to Hire

#44

Post by kleinsch »

ohazi wrote: post #29
johnrob wrote: post #27

Earlier quoted context omitted.

The co-working trial is a two way street. Wouldn't a good developer want to make sure the new co-workers are also good?

Hiring has always been a two way street, but in technology the scales are usually tipped against the hiring company. It's simply a numbers game -- there are far more mediocre and poor developers than good ones, and in the current environment, the good ones can basically get hired to work wherever they want. The co-working trial is an attempt to make the scale more favorable for the company. I can understand the desir…

I've worked at a few places that hire top level engineers and use this tactic. It seems to be very successful.

As a developer, why wouldn't you want to see what the codebase and your coworkers are like in a real environment - and get paid for doing it?

Worst case scenario: you wasted a day of vacation, got paid a consultant rate for it while still getting vacation pay from your "real" job.

Re: How to Hire

#45

Post by j_baker »

nawitus wrote: post #38

Yet another article on hiring without empirical data. And as everyone knows, "gut feeling" doesn't actually turn out to be correct that often in recruiting (otherwise we would have solved the problem already).

Do you have any empirical data to back your point up? Or are you basing your thoughts about gut feelings on gut feelings?

Re: How to Hire

#46

Post by harlanlewis »

larrys wrote: post #43
ig1 wrote: post #37

Earlier quoted context omitted.

Anytime your reasoning is "we shouldn't do it because it will make them attractive to other companies" you're probably wrong. You want a company full of super-stars that everyone wants to poach. Otherwise you're just like the companies that prohibit employees from have linkedin profiles or don't give employees training because that makes them more poachable. Crazy.

Not sure you understand what my point is. I'm not saying you shouldn't have employees that are so good others want to poach them. I'm saying that by giving someone a title (that they might not deserve) you are possibly making them more attractive and more likely to be poached. This was said simply to counter the parent point that "there is no cost". Let's say I allow you to put on your card "VP of Software Engineerin…

Or it makes them less likely to be poached, because potential poachers will discover they don't measure up to expectations after a phone screening. At this point we're all pretty used to sussing out title inflation at step 1.

Or maybe someone will perceive it in a completely different way. At an individual level, big moments in people's lives (like changing jobs) are shaped by the edge cases that lead to actions, not macro average behavior.

Making personnel decisions based on worst case hypotheticals only seems to increase the chance your employees will have a hard time imagining long term career growth by sticking with you. Deflating titles seems about as viable as forbidding conferences. Except where titles impact compensation, all of this is completely independent of actual competence.

Re: How to Hire

#47

Post by enraged_camel »

ohazi wrote: post #22

> Whenever possible (and it’s almost always possible), have someone do a day or two of work with you before you hire her; you can do this at night or on the weekends. No. Absolutely not. If a candidate really is as good as you think he/she is, they already likely has a full time job that they're perfectly happy with (as you already well know): > Often, to get great people, you have to poach. They’re never looking for…

>>Hire fast / fire fast is fine, but seeing a founder do this is a big red flag for me. It screams indecisiveness, lack of confidence, and mediocrity. Does he try to pull this stunt with everyone he hires? What am I supposed to think of the rest of the team? Can I get behind someone whose hiring process is so driven by fear that he's willing to miss out on the best candidates?

At the startup stage, someone who is mediocre or outright bad can cause so much damage that they can sink you. That is why hiring is the one area where you should be very risk-averse.

Re: How to Hire

#48

Post by sgustard »

tptacek wrote: post #25

So much interesting stuff here. For the past couple years, this has been most of my (non-billable) role at Matasano, and we've gotten steadily better at this stuff. So, my thoughts: ›››››› Spend more time ›››››› Have a mission ›››››› Always be recruiting These three are of a kind for me. I recommend that you think about hiring the way Patrick McKenzie things about conversion optimization: as a form of marketing that…

Regarding the "hang out" test. Imagine you're a person who only likes to hang out with certain people, namely people already somewhat like you, by way of gender, race, age range, grew up like you in the USA watching sports and 80s sitcoms, and so on. Now your hiring process is probably not only unfair but illegal. But your company will act as a great echo chamber to reinforce your own preconceptions, so you have that going for you.

Re: How to Hire

#49

Post by rhc2104 »

Is Sam saying that investors recommend that a first engineer gets 1.5% (and that he recommends 3%), or that he recommends 1.5%?

Re: How to Hire

#50

Post by yannisp »

"The absolute cheapest benefit you can offer an employee is their title, and if they have a rational reason for wanting a better title, why on earth would you refuse to give up some title to get a good candidate on board?"

If your role doesn't match your title, you may be in big trouble when/if your company gets acquired. Also it might seem like the cheapest in one sense but as the company grows you'll find yourself in an awkward situation when a manager makes more than a VP you just hired. It's a sensitive subject that should be approached more carefully than you'd expect.

Post reply on HN