Google tries new angle on hiring
ft.com
ft.com
I've met geniuses who don't make anything interesting out of their lives. They just play mind games and show off. Nobody really likes them or their "high standards." Usually they just criticize everybody and everything in some way or another, instead of helping out. And they are often lazy, covering themselves by stating things are not "challenging" or worthy of their highness.
They remind me of those rich guys with sports cars. They wouldn't win any race. What matters is how you do with what you have. The house of cards of arrogants falls down if enough people just ignore them. Like the typical high school hotness scenario.
My ideal hiring test:
- What do you code for fun?
- How do you help people around you?
- Favorite non-fiction books?
- Would you like to come over for a hackathon next saturday?
Google will not leave me alone for all of the times I have told them "I am not interesting in working at your company." I think they assume I am kidding.
It's my experience that if you really want to work there they probably want you.
(I'm also concerned about their discriminatory hiring practices -- selecting principally from specific schools rather than looking at individual candidates)
Google never visited my university either, but i got approached for an interview (linkedIn ftw) and i accepted it because saying no, or not applying is like turning down free lottery tickets. you might not win but it doesn't hurt.
The interviews were quite challenging (and of increasing difficulty), but there was nothing even resembling a trick question or a mind teaser. I researched a bit before the interviews, and credible results consistently said that Google always had a policy against such questions. I wonder how the head of HR can then talk about it as if they were asking such questions recently?
All in all, the hiring process was challenging and exhausting, but never felt unfair or sloppy/undirected.
Still, limiting the number of interviews to 5 or so is at the very least a positive step towards reducing the chance of people meeting someone from what steve yegge calls their "interview anti-loop" (see: http://steve-yegge.blogspot.com/2008/03/get-that-job-at-goog...; short version - sets of interviewers whose favourite subjects/questions happen to not overlap with your particular specialities, and will therefore have a much dimmer view of your talents than differently focused colleagues).
Also, a problem with many many interviews is it's even harder for a candidate who has been given no significant post-rejection feedback to figure out what needed improvement or where you weren't a good fit; I'm reduced to speculating that maybe number 7 was a bit iffy? Or maybe 4 and 7?
The common perception from people in the industry is that many of the best programmers are self taught. That doesn't preclude doing a degree of course, however even the really good programmers I've met that did degrees say that they really ramped up their learning efforts after leaving university (and that was my experience too).
Good to know they've unbent a little on that as well. Of course it then begs the question "how then do you hire good programmers?". Solving that problem would be really interesting (my pay might go up :D )
If you're smart and experienced, your degree doesn't matter much for engineering.
On your 2nd point, I totally agree with you that self-taught programmers should actually get a degree in CS or any related major. Although the programming classes might bore them out, there will be courses and professors who will take them to somewhere they would have never known before. This is just my generalization and, of course, will not apply to everyone ;)
Wow, they actually consider 5 interviews to be a lean approach to interviewing; no wonder their process is convoluted. Methinks Google seriously needs to rethink their org structure.
I work at Google and have interviewed O(250) people there.
Given the cost of a bad hire, I personally don't think that's enough.
I'd like to be able to consider it but I'm not in their candidate pool (around forty, no regular job for a few years, non-tech schooling, travel, no high tech work for years before that).
Must be hard to reach out to the kind of people Google does like to hire. My old friends from that group are always busy.
Given your situation, and the fact that going to a couple of interviews is enormously low-risk compared to the potential reward, maybe you should say yes to the Google recruiters? What is the worst case scenario here?
Also, maybe they heard (through the grapevine) that you do good work. Stranger things have happened. :D
@grief-voter: WTF dude? Care to explain?
I am not in their candidate pool, either by your standards. Quickly approaching forty, dropped out of 'society' for 5+ years, non-high tech work during that period, no Ivy League degree. And, yet, despite the fact that I have made -exceptionally- clear that I have no interest in working at their company, they will not stop recruiting me.
I think you should go for it and put in that application. It's way more likely that they want you to work there than that you want to work for them.
Don't gamble on 1/10. Most likely, you'll feel like crap for nothing. And don't listen to the 1/10ers, they often rationalize and are fooled by selection bias.
It might be selection bias, but it's not rationalization and it's not fooling me. It's annoying the crap out me. I hope they go bother someone who is interested in working there.
> Assuming you are talking about me
Nevertheless, rectangular manhole covers are pretty common.
I never claimed the market was perfect. I find that perfect omniscient markets a terrible assumption to base analysis on.
The combination of superior cover safety and superior tunnel structure are pretty compelling, but not everyone chooses the optimal solution. That's just not how humans operate.
http://farm3.static.flickr.com/2164/2377255702_fec42f19b2.jp...
If ever asked I would try to convince them of the triangular merits.
I didn't have the heart to tell him its a cliche.
* Convert decimal 11 to binary.
* How many bits are in Unix permissions.
* What does set UID mean.
* How do you rm a file named -f
* How are hard links different from soft links
* What does const mean
* More and more Jeopardy for Geeks questions
It was like a Unix/C trivia game. No reasonable person would name a file -f (but that's beside the point)...it's just fun thinking of all the ways you could remove it if you ever had to.
I've been involved in ~100 Google interviews and I've never seen questions like this in a SWE interview.
I would expect anyone who's been through a CS101-level class to do small decimal<->binary conversions in their head, and anyone who's administered a UNIX system to know how to terminate an option list.
If they have two pre-qualifying rounds of CV burning where at each stage they trash 80% of the CVs, then they'd be down to about 26 before even talking to anyone. Now four rounds of interviews where 50% of the people are eliminated and you still have 1.625 people left...
Seriously, the recruiting folks told me that the vast majority of the "blind submit" resumes that came in via e-mail and website forms were unchanged duplicates of things already in the database.
P.S. If you know why they make them round, then do you know why they don't make them triangular? ;)
First off, there will be a small lip to keep it from falling in flat. This means that the hole is actually smaller than the manhole cover by an inch or more.
Second, no matter how you orient an equilateral triangle with on point facing down the hole, there is an edge across the top that is too big to fall in. The way to visualize this, if you take one edge of an equilateral triangle and sweep it in a circle around one end, it will be outside of the triangle at all points except where the corners meet. So trying to drop it in any position other than with the edges lined up will make it impossible to fall through. And since there is a lip when the edges line up, it can't fall through.
In contrast with a square or rectangle, if you sweep one of the edges it will be inside of the square or rectangle at some points (on a rectangle, you have to use the small sides). This means all you have to do is shove one edge of the cover down the hole and it will fall in.
Exactly, and with a big enough lip/flange, ANY shape will work.
Your flange is not "big enough". For an absurd example, consider a hole 3" to a side, and a flange 48' in width.
That said, I honestly meant this as a mathematical/rhetorical statement, not a practical one. So yes, you're correct that for any reasonable flange width (given the assumption we're talking about steel/iron manhole covers, and accepted values of the material's strength, etc.), some shapes wouldn't qualify.
My point was to show that the question, taken at its face value, isn't even a good question. It assumes a flange, or no shape would work as a cover, as the cover would simply fall into its own hole. Thus, if we're to assume a flange (or a taper; it could be argued wine bottle stoppers and the holes they "cover" are both "circles"), we should be able to assume one of arbitrarily large size, and then ANY shape will work.
> Because all diameters are the same length, the Reuleaux triangle,
> with all other curves of constant width, is an answer to the question
> "Other than a circle, what shape can you make a manhole cover so that
> it cannot fall down through the hole?"
Soon, they'll put shit (err-- a chip) in your brain, then follow you around for 6 months.
"Its just a trial."