You Think The Google Interview Is Tough? Try Getting A Job At A PE Firm
businessinsider.com
businessinsider.com
I've never been a fan of this kind of blind arrogance in the finance industry. I hoped the recession would at least have some effect on that, but it probably won't. Engineers don't have nearly enough power in this society.
I know for a fact that most public accounts of Google interviews are wrong. I have no reason to believe that public accounts of financial company interviews are right. I just assume that everything I read on the Internet is wrong until proven right. ;-)
FWIW, my Google interview was much harder than my two interviews with financial software startups. But I'm not certain if those startups are representative of the financial industry as a whole. Things might be different at Goldman, or hedge funds, or PE.
The simple truth is that people like the interviewee in the article just don't work on nearly as hard of problems as do Google engineers. Being an engineer myself, I've never had a finance job, but I've had plenty of friends and roommates who have. They need to know some math and have some Excel skills, but most of their time is spent on phone calls and Powerpoint decks. That's why even an easy first question in an interview for an engineering position (e.g. write a C function that reverses a string) is more difficult than the questions asked of these guys. Just look at the example of the most grueling question in the article. Describe the process of a leveraged buyout? That doesn't really require any analytical skills, creativity, or problem solving.
I really think it's suboptimal to have finance people running our businesses, Wall St., and our economy. In my experience, they have a hugely inflated sense of self-worth, which is based mostly on their hugely inflated salaries.
Out of interest, how do you know this for a fact? I presume you were actually at most of these interviews?
The best hires are always the people that you've worked with in the past, or who come personally recommended.
Google's dick-swinging about its interviews is IMO not really one of their strong points. The interview process can eliminate some complete losers, but it probably also eliminates lots of people who would have been great on the job.
I have some empirical backing for this. During my brief tenure at Google our team worked with a lot of interns, and we made our recommendations for extending offers. Then, suddenly someone made a new rule that outgoing interns should be re-interviewed by members of a completely different team, I guess, for objectivity. And a lot of them were rejected through that process.
So we had people who had proved themselves in an initial interview, and during many months of on-the-job contributions, getting rejected because some wanker asked them some question about missionaries and cannibals or some other useless puzzle question.
And yeah, it was that sort of thing -- I remember one intern trooping dejectedly back into our offices and I and the tech lead tried to figure out what the hell the answers were to the question(s) they got asked. We couldn't figure out those puzzles either.
Why would someone with a 4.0 in a hard, rigorous subject go into engineering for a measly 60k a year starting salary + 20% raises when he can work in a finance industry where the top talent makes in the hundreds?
I had a lot of friends in the math department that were double business majors. I imagine that at least for quantitive finance positions, interviews such as these (if not way harder) are a must to separate the top talent from the elite talent.
A trader at a top tier firm, on the other hand, might have a $150K+ base first year out of school. Adding on bonuses, it isn't unusual for them to be making $1mil/year within 3-5 years (although the vast majority don't make it five years ...). A quant programmer (slightly less competitive/stressful) can expect to be making $200K/year+ within five years at a top tier quant firm.
It isn't really fair to compare programmers to bankers since the hours aren't really comparable. But the traders/quant-programmers I know (usually) only work about 60 hours a week.
All based on my relatively limited experience (~5 people/company - although I have no reason to suspect bias in my sample).
I'd love to hear what a good answer to that question is. Anyone up for it?
"Shit's Easy Syndrome".
You know. As in, shit's easy. If it's easy to imagine, then it's easy to implement. Programming is just turning imagination into reality. You can churn through shit as fast as the conscious mind can envision it. Any programmer who can't keep up is an underperformer who needs to be "topgraded" to make room for incredible new college hires who can make it happen, no matter what "it" happens to be, even if they have to work 27 hours a day, which of course they can because by virtue of being new college hires, they have no social lives and no spouses or significant others, and they probably smoke a lot of crack from being in the dorms so they can stay awake for weeks at a time.
So there are eight million people in NYC. Let's say that each household is two people, on average. So four million households. Because NYC is crowded, not many people have pianos in their homes -- let's say one in fifty. So 80,000 pianos. Let's double that to account for the schools and concert halls, so 160,000 pianos. On average, let's say a piano is tuned once every other year. So each year, there are 80,000 tunings.
A piano has 88 strings, and let's say a trained tuner could tune about two per minute, so a piano takes forty-five minutes to tune. If the piano tuner knows where he's going, and makes appointments accordingly, let's say it takes another forty-five minutes to get from one appointment to the next. So ninety minutes is one piano tuning. One tuner works 50 weeks a year, 40 hours a week, so 2000 hours, or 1333 tunings. So 80,000 tunings / 1333 tunings per tuner = 60 tuners in NYC.
Of course, that assumes each tuner works full-time. If we assume there are part-time tuners, that number only goes up. I'd round to 100.
You see, the point isn't to get the exact number; the point is to show that you can figure out impossible problems.
The typical background for someone applying for this sort of job would have 5-10 years spent at a management consultancy firm, maybe a few years at an investment bank and an MBA. They'll have read thousands of business case studies and written dozens or hundreds themselves.
They'll have seen a number of LBO/IPO deals and will have spent hundreds of hours studying how these are structured.
These are the kind of people who when they read about a company automatically start thinking about market sizing and modelling.
Not being able to do a case study in an interview would be like a programming interviewee not being able to solve a trivial problem like FizzBuzz.
And as for the timing, it's just testing something that anyone with the appropriate background should be able to breeze through. If you're meeting with a C-Level client, running over your allocated time is incredibly rude, if you can't manage your time in an interactive discussion situation then you don't belong in the industry.
The whole article just sounds like it's based on the rants of a candidate who wasn't capable of doing the job they were applying for.
Financial professionals whining about how much it takes to make the wheel turn in their industry strike me as about as honest and reality based as dressage shops for unicorns trying to make a case for why their particular brand of mastery ought to sit at the top of human endeavour.