Quants are the new rock stars of finance, straddling a brave new world of big data and machine learning as computers take over. If you want to get in on the action, though, you’ll need to know how many ping pong balls fit into an Olympic-sized swimming pool.
‘Trick’ maths questions (since ditched by Google), marathon modelling sessions and an extreme vetting process – welcome to the world of quant recruitment. Investment banks and hedge funds don’t just want to hire any PhDs, after all, they want the best and the interview process can be gruelling as a result.
“Last year I had a seven hour non-stop marathon session for a model validation role with Goldman Sachs onsite in New York, where I was interviewed by eight people,” says one quant VP at a U.S. bank in New York. “Towards the end, I was so exhausted that I did not perform as well as could have.”
The most common complaint among quants is that the whole interview process is designed to trip them up. One tells us that a bulge bracket bank asked them to write down the Black Scholes Formula as the first question in the first interview, another Maths PhD says that he a recruiter asked “deep computer science” questions that were irrelevant for the role they were applying to.
Jonathan de Montfort, a former Goldman Sachs quant who now runs his own blockchain and algo trading start-up, says that there’s too much of an emphasis on “trite questions” or theoretical mathematical concepts that have no practical application in financial services.
“They sometimes also ask specific mathematical questions like an integration or differentiation,” he says. “But they’re all the same. Knowing such things doesn’t mean that you’ll be able to apply that to what we do in finance.”
For all the talk of hedge funds and investment banks scrambling to hire quants, actually getting the job is something of a lottery. Jamie Walton, the former head of FX quants at Morgan Stanley, says the bank received applications from 700 quants every year, and hired 10-20 of them. Successful candidates would go through three rounds of interviews, meeting around 10 people each time.
“It’s not like having a PhD is a guarantee they’ll be a good hire. Someone might have studied the Monte Carlo methods in finance, have all the right background, but then you interview them and realise they know nothing,” he says. “On the other hand, someone might have a PhD in astrophysics, but have all the qualities banks are looking for.”
Walton says that quant interviews aim to test three things – mathematical ability (hence the trick and hard maths questions), coding skills and, more importantly, how much PhDs who might have no banking background know about the financial sector.
“Investment banks have to be thorough, because a bad hire can be disastrous and a really good one makes a big difference,” he says. “It’s about finding a diamond in the dust.”
De Montfort believes that banks are still getting it wrong by focusing too much on mathematical skills rather than practical know-how.
“At most you need one or two pure maths PhD type people – experts who might help with some level of understanding on specific mathematical points when needed,” he says. “Quite often though, the maths experts would not be especially good at understanding the vagaries of trading, which cannot be quantified so easily and require some level of judgment.”
Once quants have industry experience, very often they “forget simple concepts” they learned at school, says the quant VP in New York who worked for Morgan Stanley and Citi. Still, banks continue to test for everything in interviews.
“Quants need a lot of different skills – maths, statistics, programming, finance – so there is a very broad range of questions that can show up during interviews,” he says. “Candidates are usually not given any information as to what to expect. Nobody can remember everything they learned from school.”
Then, there’s the dreaded feedback loop before the green light is given to a new hire. The VP quant says everyone involved in the Morgan Stanley recruitment process has to say ‘yes’, ‘no’, or ‘maybe’ on a potential recruit. One ‘no’ means they’re out of the running.
“At Morgan Stanley I have seen candidates who received four positive feedbacks, but one negative feedback from a computer science PhD, and they ended up being rejected,” he says. “In the end, they hired a candidate who everybody was ok to hire, but ended up being a disaster for team chemistry.”
Walton agrees that everyone has to give some sort of thumbs up before a quant is hired, but insists that it’s not as clear-cut as that. By the time quants are invited back for a third interview, the ball is in their court, he says.
“By this point, it’s more about due diligence and we’re pitching the bank to you,” he says.
Still, some quants we spoke to think that a better solution would be undertake a “technical dating” process where candidates and potential employers work on a short-term two-week project together where they get to assess mathematical competency, finance experience and softer ‘fit’ factors. But with 70 applications for every job, maybe banks are too busy to do this.
Photo: Getty Images