Hello,We are hiring for a junior quant(-ish) role. As they'll be working with me I've been asked to interview them, however I'm not a mathematicianand really have no idea what sort of questions I should ask in order to filter out bluffers from those with potential.This is the first time we're hiring for this sort of role (they'll be quant #1 in the company), all we know is that they'll be required to come up withpricing models, will need to know about 2nd price auctions and more be able to do more general analytics.So, any questions people can suggest would be greatly appreciated.

Q1. When did you first learn about 2nd price auctions? My answer would be: about 2 mins. ago. (If you are looking for some deep knowledge of auctions, you're asking forsomething well outside typical 'quant knowledge', IMO). Q2. Can you explain Black-Scholes hedging to me in very simple terms?Good answer: they will talk about a stock that can only go up 50% or down 25% in one period (or something similar), draw you a diagram, perhaps write a couple of simple equations using high school math, and after a few minutes -- you will almost understand it.Bad answer: they will start talking about stochastic calculus, or partial differential equations, and you will be immediately lost.Q3. Here is a relatively simple project typical of our company. Take it home, think about it, and write up a couple page answer for us.One for you: what does your shop do and what kind of pricing models?

Last edited by Alan on March 3rd, 2016, 11:00 pm, edited 1 time in total.

QuoteOriginally posted by: AlanQ1. When did you first learn about 2nd price auctions? My answer would be: about 2 mins. ago. (If you are looking for some deep knowledge of auctions, you're asking forsomething well outside typical 'quant knowledge', IMO). Q2. Can you explain Black-Scholes hedging to me in very simple terms?Good answer: they will talk about a stock that can only go up 50% or down 25% in one period (or something similar), draw you a diagram, perhaps write a couple of simple equations using high school math, and after a few minutes -- you will almost understand it.Bad answer: they will start talking about stochastic calculus, or partial differential equations, and you will be immediately lost.Q3. Here is a relatively simple project typical of our company. Take it home, think about it, and write up a couple page answer for us.One for you: what does your shop do and what kind of pricing models?Alan, thanks for your input.We're in the real-time advertising business. The whole of this industry works around 2nd price auctions.The money's probably not as good as finance, but the work is very interesting and the work/life balance is unbeatable.

- Cuchulainn
**Posts:**60470**Joined:****Location:**Amsterdam-
**Contact:**

I think Alan's Q3 is a good one. I don't know how much skills are expected but you could ask to create a bare-bones prototype (in Python or C++(?)) based on your financial model and then discuss it at a detailed second interview. Then you can quiz him/her on the what-if scenarios based on the prototype.Indeed, I would avoid too much discussion on generic quant maths. Would a quant be a better fit than say someone with strong CS skills who is good with maths?I heard of cases when quant newbies had a week to study the book by Hull and in the second interview they would be grilled on it. Maybe something similar for auction theory? My gut feeling says it is a complex state machine?

Last edited by Cuchulainn on March 4th, 2016, 11:00 pm, edited 1 time in total.

http://www.datasimfinancial.com

http://www.datasim.nl

Approach your problem from the right end and begin with the answers. Then one day, perhaps you will find the final question..

R. van Gulik

http://www.datasim.nl

Approach your problem from the right end and begin with the answers. Then one day, perhaps you will find the final question..

R. van Gulik

QuoteOriginally posted by: CuchulainnIndeed, I would avoid too much discussion on generic quant maths. Would a quant be a better fit than say someone with strong CS skills who is good with maths?I heard of cases when quant newbies had a week to study the book by Hull and in the second interview they would be grilled on it. Maybe something similar for auction theory?Exactly. I think the challenge for the advertising company is to find a good, technically-oriented PhD, and then convince them that thebusiness contains enough substantial problems to be worth their time. Given that, you can teach them enough about the auction business to get started,and a PhD should be able to take it from there. Certainly 'quant-refugees' will provide a pool of candidates (and presumably the reason the OP is posting here),but also fresh graduates, say with Econometrics, Statistics, Applied Math, or CS degrees.

Last edited by Alan on March 4th, 2016, 11:00 pm, edited 1 time in total.

- katastrofa
**Posts:**8494**Joined:****Location:**Alpha Centauri

"Q3. Here is a relatively simple project typical of our company. Take it home, think about it, and write up a couple page answer for us."Isn't there a threat they will discourage the best candidates (who have a job already and thus don't have time)?dd3, purely out of curiosity, would your company consider outsourcing the work on building your trading models?Regarding the original question, I would begin by asking the candidate about their PhD/MSc. Then I would tell them what sort of problems there are on your plate and ask how they would approach/try to solve them.

Last edited by katastrofa on March 4th, 2016, 11:00 pm, edited 1 time in total.

QuoteOriginally posted by: AlanQuoteOriginally posted by: CuchulainnIndeed, I would avoid too much discussion on generic quant maths. Would a quant be a better fit than say someone with strong CS skills who is good with maths?I heard of cases when quant newbies had a week to study the book by Hull and in the second interview they would be grilled on it. Maybe something similar for auction theory?Exactly. I think the challenge for the advertising company is to find a good, technically-oriented PhD, and then convince them that thebusiness contains enough substantial problems to be worth their time. Given that, you can teach them enough about the auction business to get started,and a PhD should be able to take it from there. Certainly 'quant-refugees' will provide a pool of candidates (and presumably the reason the OP is posting here),but also fresh graduates, say with Econometrics, Statistics, Applied Math, or CS degrees.So some of the tasks they'll be involved in will be pricing and arbitrage. The 'challenge' with 2nd price auctions is that you don't know the value of something until you're the one who wins the auction. We would have to come up with a model that lets our business deliver client requirements at a low cost as possible.Again for arbitrage it's a similar problem... try to sell to a 3rd party at a high enough price to pay for the IT cost and to increase the probability of us being the final winners of the auction, but to set a price low enough that the 3rd will actually enter the auction.We currently participate in about 15 billion auctions a day and we typically have a maximum to 200ms to analyse an auction and if we decide to participate, calculate a price.I believe we're trying to get fresh grads, it's not up to me.

QuoteOriginally posted by: katastrofa"Q3. Here is a relatively simple project typical of our company. Take it home, think about it, and write up a couple page answer for us."Isn't there a threat they will discourage the best candidates (who have a job already and thus don't have time)?dd3, purely out of curiosity, would your company consider outsourcing the work on building your trading models?Regarding the original question, I would begin by asking the candidate about their PhD/MSc. Then I would tell them what sort of problems there are on your plate and ask how they would approach/try to solve them.Hi katastrofa,Outsourcing is definitely not an option.

If you are not a quant yourself, why beat yourself on it . This is a junir quant , so just ask some aptitude base questions or puzzles. Or just ask some methodology related questions like put-call parity, products based questions and basic pricing like IRS and Bonds.

Because I'd be asking questions I don't know the answers to.

Anyway, we hired a guy from some credit risk place, and he's underworked and overpaid. Nice chap though.

Anyway, we hired a guy from some credit risk place, and he's underworked and overpaid. Nice chap though.

that's a pretty good job title(they'll be quant #1 in the company)

Ask them about your field. If you're not a mathematician they will very quickly learn you know nothing about models and play with you. I did it a couple of times.Hello,We are hiring for a junior quant(-ish) role. As they'll be working with me I've been asked to interview them, however I'm not a mathematicianand really have no idea what sort of questions I should ask in order to filter out bluffers from those with potential.This is the first time we're hiring for this sort of role (they'll be quant #1 in the company), all we know is that they'll be required to come up withpricing models, will need to know about 2nd price auctions and more be able to do more general analytics.So, any questions people can suggest would be greatly appreciated.

GZIP: On