Remember that time I told you not to go to 4chan?

This is where I tell you that you should have gone to 4chan. Because then you'd know for sure what the accurate starting salary of a math PhD was. According to a meme.

Well, we can all cross that bridge when we come to it I guess.

I don't really know that much what I want, but if I can avoid it I'm going to avoid waged labor. Or are Google's engineers salaried? I'm not sure, but I'd rather earn money independent of the pocketbooks of some corporation.

Also, that ****ing uncleanable bumpy texture they put on appliances now. It's to cover for the fact that spray paint doesn't have a uniform thickness like the powder coating did. It's literally visual noise they added so you can't instantly tell how their production quality fell off a cliff.

Google never really struck me as the kind of company anyone with dignity would work for.

Not saying I have dignity, there's an almost certainty I'll sell out, I just hate the idea.

I don't know about their dignity, but there is a large contingent of researchers who come from academia, Bell Labs, MIT, PARC, and DEC who go to Google in order to be around each other, and presumably to work on large-scale problems in distributed computing / A.I.

I bet it's a fine place to be intellectually, it's more that you'd be working for Google, which is basically cyber Hitler.

And if that makes the USA Nazi Germany, would you be prepared to leave?

Amerika, love it or leave it.

Some people just wanna eat the **** sandwich, dude. $250k total comp to do some ad tech ****, guaranteed retirement in 30 years, that is not a bad deal if you're the kind of worker who'd just be running out the clock anywhere else.

If you do decide to sell out so that you can take the money and run, just make sure you read the fine print. A big chunk of your compensation will be in the form of stock options which don't vest fully until you've worked there for four years, and quitting before then will significantly reduce what you ultimately take home.

True. I'd consider it. Early retirement can be very nice, it gives you time to live life a bit.



I mean, it's hard to live elsewhere when you're monolingual and monoculture, but yeah, I wouldn't mind leaving America someday.

The forecast over Antarctica is looking pleasant.

I don't know about the status of their technical fellows. But I have heard that Zuckerberg pretty much Florida Man'd Google's face off a few years ago, poaching a ton of their A talent from the glory days. Then a lot of Ballmer era Microsoft employees replaced them.

This is all second hand knowledge, keep in mind. Dunno how true it is.

^ also repayable signing bonus, repayable relocation package. Last stats I saw were that the average tenure at Google is just over a year (people stick around until their relocation lock in ends and nope the **** out as soon as they can afford to)

That's interesting. One factoid floating around HN and the blogs these days is that Google's infamous whiteboard interviews have gotten a lot easier.

There also comes a point when experienced folks refuse to put up with a dehumanizing interview process that is rife with false negatives and consists of months of shenanigans by disinterested programmers checking off boxes in quiz problems.

The problem with the Google interview was never that the questions were too hard, because they weren't.

The first problem was that their interview process was designed specifically to remove social interactions from the equation (i.e. hire smart *******s). It worked too well.

The second problem is that they put their engineers personal feelings about hiring ahead of their own data. Google's own data showed that their interview scores had zero correlation with job performance. Zero. It was as good as doing nothing at all - actually, worse, because it wasted the valuable time of their highly paid engineers. Despite this data, they never really did anything to reform their practices. It was too socially important to Googlers to change, a kind of "I had to put up with this bull****, so do you" or the even more toxic, "this process successfully identified the magnificent super genius, me, and therefore it is working perfectly and should never change". It's really hard to change this kind of repulsive machismo **** once it's in your culture.

I don't believe for a second that Google has really changed anything here. Every couple of years someone posts something to that effect, it always turns out to be bull.

As far as I can tell, most everyone everywhere has no idea how to hire. Except Hooters.

Software hiring is impossible. At this point I'm pretty close to just giving candidates a typing test, because at least that can tell me whether they've even seen a computer before.

Can I work from home?

can you touch type {, }, &, and *? Then you've probably written code. Welcome aboard.

Wait, do you really get people who are completely inexperienced in programming applying?

Oh, no, you misunderstand. I don't interview juniors.

I interview principal engineers

I'm going to sue you for discrimination.

I have turned away ~famous programmers~ because they cannot follow directions or ask questions.

And really, I try. It's my personal goal to humanize the process as much as I can, explain exactly what's expected of them and provide immediate positive feedback when they're moving the discussion in a direction that is useful or interesting. Everything I can do to account for stress. And I made a standard rubric, which candidates can see.

And still.

Regexp facing 20 years for its discriminatory pattern matching ways, news at 11

My head's going to explode, what's wrong with the world where this happens?

I am really kicking myself rn because I can't find it, but I could have sworn I read somewhere that Amazon tried a pilot program where they completely took human interviews out of the process of hiring new grads.

****ed if I know.

Please elaborate. What are your thoughts about the problem here?

Speaking of taking humans out of the process, is that you ELIZA?

So I'm 2.5 chapters into an economics 101 text, and besides the review all I've gathered is that supply/demand curve economics is just easy optimization with some tacked-on explanations

Found the Amazon thing. Can't speak for its veracity.

Well, I guess that explains why Amazon continuously rolls out inferior hardware/software packages.

This is the "best" interview process I've developed so far.

- Phone screen, ask a simple counting problem to make sure they aren't morons, shoot the ****.

- 3 hour work sample, graded on standard rubric (this replaced other phone screens and in person interviews when I added it)

- Maybe a follow up analytic problem if they're borderline.

Then on site:

- Beforehand, do a thorough code review of the assignment. Then give the candidate a barebones VM they need to set up, build their assignment, and respond to code review feedback.

- Then team fit interviews.


Or, at least, I hope this will work. Just as soon as we get a work sample that compiles.

****ed. If. I. Know.

I shouldn't be nitpicking economists, but I wish they would use language that's compatible with mathematics and not.. this.

This may have more to do with Jeff Bezos being an utter and complete control freak. Fire phone was a disaster but Bezos went full steam ahead to the bitter end.

Apparently, the man still has the final say on every single pixel of Amazon.com.

I feel like asking someone a question which embeds the K_{3,3} graph and seeing how long it will take them to realize it's impossible to embed on a plane would be a good filter. Or maybe I'm optimistic? I want to help even though I know nothing about hiring.

There are axiomatic treatments of econ that try to be "mathematical" in a more rigorous sense. Not really sure how much of a cargo cult this is, or more to the point, whether or not it's worth your time to go down this road (not sure if I can emphasize this enough, don't let rigor-mortis hide the elementary utility of the theory be hidden by abstract non-sense), but I've heard good things about Ariel Rubinstein's books, which he has available for free as PDFs on his site. I've glanced through his micro book, which seems to be geared for grad students, but the prose seems to be good, and it uses precise mathematical language that you might appreciate as a mathematician.

Some other reasons I think he might be a good author:

1. There was an Amazon reviewer who was a fan of Ayn Rand and got super pissed that he talked about his socialist views in the epilogue of one of his texts
2. He basically thinks that economics is no more than a collection of fables, and is highly skeptical that game theory can possibly be useful, except as allegories--despite the fact that he himself is a game theorist and has published a book on game theory

Disclaimer: I don't even know econ 101, so don't listen to me. But if I had to hold my nose and learn it, this is the route I might take, with the qualification that it's worse to be fooled by fake mathematical rigor than read an econ book that uses "sloppy" math but gets all the important basics right.