[FoRK] what the hell, facebook
sdw at lig.net
Wed Jun 19 01:57:23 PDT 2013
On 6/19/13 12:42 AM, Reza B'Far (Oracle) wrote:
> Well, or that he flipped off the interview process ;-)
As partly a favor to a friend, I did finally interview at Google a couple(?) years ago. I was overexcited about it, but it went
well. I was told that I "passed" the interview (8 people on 2-3 different days), and was asked to meet with three teams to
decide which I wanted to work with. (After being initially earmarked for and then discarded by Project X. I would be perfect
there for various reasons, but not yet useful with machine learning at that point. Daphne was my first brick wall ever.
(-Daphne, +Thrun, ++Ng. And I will leave it at that.))
I was and remain conflicted because I have some things that I want to do on my own before I get sucked in to one vortex or
another, but I would have fun at Google and likely wouldn't have or feel problems that some have there. Although I was earning
nice consulting dollars at the time and didn't know then what they could pay / bonus.
I probably would have been weak at the time for various reasons and taken something interesting, just trying to work out
something for this set of ideas I'm nursing.
However, I was saved from that fate because A) the young leads for one group stated that I was too old for them and B) someone
in the hiring committee overruled the other groups because someone felt I was arrogant.
My guess is that I was too honest about my goals, or that I felt that at a couple fleeting moments that I was in the rare
position of being one of a small number as authority on something, or maybe it was that the Android dev in charge of the new
garbage collector took offense at my suggestion that I seemed to have proved that it had a problem garbage collecting circular
referenced objects with no other references. (I sent code later.) See the link below. That code sample also shows a Dalvik bug
that caused bade code generation, fixed by renaming certain variables.
Or, I'm just obviously and hopelessly arrogant. You know what I always say at this point: Don't confuse arrogance with chronic
correctness. Of course, you only get to that point by trying many incorrect paths, at which point you then are an authority on
how not to do things. Probably more valuable than immediate success. Just a recapitulation of Edison's 5% inspiration maxim.
But, you know, all those long nights knocking my head against the keyboard to solve often dumb problems that I created have just
made me arrogant. Is it arrogant to have confidence that, given enough time, I can plow through most problems to success,
having done it a few times? I don't think that word means what you think it means. Am I going to keep razzing Google about
these two questionable calls (one of them illegal) until they some-day give me an offer I can't refuse? Probably; I'm
unrequited. They're big boys and girls, they can take it.
But that will have to wait, I'm deep in a hot startup doing cool work with lots of interesting career upside, especially the
chance to work on many applications of machine vision, machine learning, photography, imaging, 3D, mobile, and many fields and
applications that could use better sensors and algorithms. With a lot of smart people that I and others are hiring, access to
interesting companies and technologies, and very difficult technical, management, and product problems to solve. There are few
positions in aforementioned vortexen that would be as fun, challenging, or rewarding and I'm not likely to land in those
positions. None of them would simultaneously use many different facets of my experience like a small, complex, highly technical
startup does. I go from reviewing / participating / writing designs, features, different types of algorithms, to
team/service/offshore management, sysadmin, contracts, policy, photography, relationships, and humor in a single day sometimes.
Our latest promo:
It would be interesting to interview again some time at interesting companies, although I won't have time to do that for a
while. I've now reviewed hundreds more resumes, phone screened something like a hundred people, interviewed many dozens, and
given offers to many. I've learned many interesting things about other companies, products, projects, and developers of various
types. I've interviewed more than one person from each of 10+ different projects at large companies. I might soon interview
someone who worked on code that I wrote.
I turned one old idea I had years ago for a programming contest problem into a Socratic interview question that gets all kinds
of interesting responses and is a highly revealing index of problem solving and programming. Hard to find something as perfect
for a backup question.
> On 6/19/13 12:35 AM, Noon Silk wrote:
>> On Wed, Jun 19, 2013 at 5:31 PM, David Edmondson <dme at dme.org> wrote:
>>> On Wed, Jun 19 2013, Stephen Williams wrote:
>>>> On 6/18/13 10:53 PM, Reza B'Far (Oracle) wrote:
>>>>> Sweet!!!! What's the URL? ;-)... props either way.
>>>>> On 6/18/13 10:45 PM, Joseph S. Barrera III wrote:
>>>>>> On 6/18/2013 10:32 PM, Reza B'Far (Oracle) wrote:
>>>>>>> FWIW - Google employees commenting on Facebook is as impartial as
>>>>>>> Facebook employees commenting on Google. May be we can make a
>>>>>>> static A.I. rule for the open world massive all-knowing
>>>>>>> A.I. engine Google will build? :)
>>>>>> "Will" ?
>>>>>> JUST KIDDING
>>>>>> ha ha
>>>>>> oh dammit
>>>> [Do I want to be on the inside working on the engine? Or outside using
>>> & building on the engine?]
>>>> This line of thinking seems familiar...
>>> Can one of the Googlers here just send Stephen an application form,
>>> pronto, please? The public courtship is getting embarrassing :-)
>> My guess would be they've already tried to hire him.
More information about the FoRK