"CV's (people lie) and reference checks (everyone knows 2 people they like) "
Marissa Meyer was asked on the Charlie Rose show about Google's hiring process, and she answered that they had done the numbers and the only correlation there was between the hiring process and how well new employees worked out was their CV. People that had prior experience in the area they were hired for did better than people who didn't. Except for that there were no significant correlations to be found.
I believe Marissa Meyer over Zaki Mahomed(the author of the article)
You have a good point and in my (much more limited than Ms. Meyer) experience, the CV is a very strong indicator. But there are times it does not work:
1. The candidate has to be truthful or you have to actually verify. Also, its worth being very careful about exactly what is said. I have met someone who genuinely had 4 years as a Senior DBA. It was true, and his former boss loved him. But when I proved for details, he was the senior DBA because he was the only DBA in the small company and mostly maintained a system consultants had set up earlier. He did not seem ready to do new development or manage a major system to me after a thorough interview.
2. The candidate is straight out of school. While there are exceptions, most people don't have much of a resume at that point.
3. The candidate is making a career change. Similar to being straight out of school (in fact they may be straight out of school for the second time) they won't and can't have a lot of experience in that particular field.
Of course, 2 and 3 might indicate you don't want to hire them for a senior position anyway, but it does make using the CV as the primary determinant challenging.
I don't think he means that the CVs are complete fabrications, but rather not wholly accurate.
His suggestion is to be applied on top of the traditional gauges of qualification to reduce the margin of error so that the truly spectacular individual can gain an edge over the noise caused by exaggerations.
So, the "best hiring tip ever" is a question that's easy to bullshit once you've heard it, and that puts people whose proudest achievements relate to their careers in the uncomfortable situation of inventing a response that dresses up the importance of their hobbies? Brilliant.
Advice like this makes me realize why employers believe that qualified applicants are so hard to find. If this is the garbage that passes for best practice in interviews, they might as well be sacrificing chickens to the gods of employment.
I'd say it's a very easy to verify question. If someone is experienced in interviews (you can usually tell that's the case) they might talk a lot about the stuff they did, but they will have nothing to show. Usually when you're excited about something you will have some proof, or will know enough details that it's easy to confirm. For almost all people I work with, it's enough to google their name to find their projects / blogs / ... everything the employer would be interested in.
Now you're hinging your hiring decision on your ability to verify a person's hobby?
It seems to me that it would be both simpler and more effective to spend your energy trying to determine if the candidate can do the job for which you're hiring. Leave their hobbies out of it.
Of course. I didn't say that is a good question, or that the answer should be taken into consideration when hiring. I just don't agree that it's easy to talk BS in that situation if the person asking it really wants to know the truth. (i.e. they don't ask it just because it's on the list)
Personally - I think it's a good question to decide which person you'd rather spend half the day with, if you have 2 equally good candidates.
Buzz off. How is this at the top of the front page on HN?
Also, it's idiotic to to perpetuate the "rock star programmer" meme. All the great developers I have ever known would never consider themselves rock stars, however most of the biggest assholes I've ever known did consider themselves rock stars. As a developer, I'd never work somewhere that looked for rock stars, ninjas, or what have you. It clearly shows they don't understand programmers and programming.
Yes I read your post. That's why I was able to quote from it. If you want to hire a good programmer, ask them about programming. It's that simple. Set up a few scenarios and let them talk and whiteboard their way out of it. Ideally you'll be able to come up with a range of problems that range from familiar to totally unknown. The familiar scenarios will tell you if they are bullshitters, and the unfamiliar scenarios will tell you how quickly they can adapt to new environments.
While it is true that people who program outside of work tend to be good programmers, it is not true that all good programmers program outside of work. And your note about charity work is outright insulting to anyone who has spent their free time helping those in need.
I think there's a lot of validity in the assumption that if you are extremely technically capable (a good programmer), then it is also extremely unlikely for you to not have come up with an idea (and thus be passionate about it) and committed to its realisation. That is, if you truly are good, it's pretty unlikely that you never implemented something on your own.
But also keep in mind it's a tip for hiring into startups, and startups at the early level require a certain amount of passion that usually comes with programmer who actually love to program (and if you love to program you've probably done it outside of necessity before).
my comment was not meant to be insulting to those. It was meant to insult people who tend to blow hard and do little real good. There are lot more of those out there, sadly. No offense intended.
I just got done doing a 4 hour long programming problem as the first start in the interview process at a company. In my long history of finding jobs, this is the way to go. It was a well thought out, challenging problem, and the solution really flexed a lot of different aspects of what a good programmer should be able to do.
It has the added bonus of me thinking this might be an above average company and I'm more interested in them than any of my other prospects.
My experience has consistently been that the only reliable way to tell how good someone is is to work with them on a real project. Since that actually works, and nothing else does, I'd say it's the interview and hiring processes that are broken.
would there be a way to try out a new hire for a 'trial period', maybe on like a 3-month starter project paired with someone more senior? then after the new hire has proven him/herself over those months, a formal offer could be made.
There are ways to do that, but there are a lot of complications.
First, most truly good workers, especially in the tech industry, aren't sitting idle for a long time, so they are often either being hired away from a job they already hold and possibly being asked to relocate. A lot of those good people will not make a commitment like that to you and disrupt their lives if you aren't ready to make something of a commitment to them. So, this might work well for entry level positions but probably less so for senior people.
Then there is the cost of dealing with them for three months or until you figure out if they are right. While this is much less expensive than keeping the wrong person on for a long time, it is still very expensive. Most managers want to try very hard to find the right person the first time not go through a whole series of short term hires looking for the right person.
In some situations this is a great idea (and what many companies actually do), but there are many occasions it is impractical.
When I started at my current (also first post-college) job, I was a contractor for about 8 months before becoming a regular employee. Two of my coworkers have made the same contractor->FTE transition. When I started it was explicitly a contract-to-hire position... for whatever reason we apparently don't do that any more, although our latest full-time position was first offered to some of the current contractors before our boss started giving us resumes to look at.
I could definitely see this reducing the applicant pool, although a surprising number of people have been looking for hotels on their first day, and then live here during the week and drive several hours home on the weekends.
Meta point: There's mixed views on HN'ing your own stuff, but I think it's pushing what's "decent" to use such subjective headline as "The Best Hiring Tip".
Even if that's the title of the post on your own blog, I think most HN'ers enjoy a bit of objectivity.
You probably shouldn't even have to ask a passionate person this; at some point in the interview, they probably would've told you what they enjoy doing outside of work.
Regardless, there's no perfect way to interview someone for a development job. It's equal parts experience, your feeling about them, how they fit into the existing organization, and a ton of other variables.
In my experience the "Best Hiring Tip" is actually: ask the person to do the thing that you are hiring them to do (or a representative exercise), and see how well they do it, or how much potential they demonstrate.
Marissa Meyer was asked on the Charlie Rose show about Google's hiring process, and she answered that they had done the numbers and the only correlation there was between the hiring process and how well new employees worked out was their CV. People that had prior experience in the area they were hired for did better than people who didn't. Except for that there were no significant correlations to be found.
I believe Marissa Meyer over Zaki Mahomed(the author of the article)