I applied through an employee referral. The process took 2 weeks. I interviewed at Knight Capital (Jersey City, NJ) in Aug 2012
Interview
Two technical phone interviews then in-person interview. On the first phone interview, interviewer asked about randomly positioned points on a circle, then to step through an analysis of Buffon's Needle. On the second phone interview, interviewer asked about a gambling game which was sort-of inverse Prisoner's Dilemma with repeated playings
Interview questions [1]
Question 1
Buffon's needle. I was familiar with it and still had a lot of trouble walking through the analysis.
I applied through a recruiter. The process took 5 days. I interviewed at Knight Capital (New York, NY) in Jun 2012
Interview
Two interviews. First one with their recruiter, talked about some general things and why I'm interested in finance. The second one was an all-round phone interview with algorithms, coding and probability questions. In the end they did not think I fit the position well.I did not have the calculator handy to calculate the expected reward exactly (see below).
Interview questions [1]
Question 1
How much would you pay to play this game? You have a fair coin. You get heads you win $1 and can continue to play. You get tails, you collect your winnings ans stop the game.
Second question. Now what if the winnings double each time you get heads.
I applied through university. The process took 4 weeks. I interviewed at Knight Capital (New York, NY) in Nov 2011
Interview
I was first sent a math/programming test to solve from home. After passing the test I was interviewed three times over the phone. I got rejected after the third interview. I thought the written exam was fairly complex, but doable. The first two interviews went well, but the third one did not. I couldn't understand what the interviewer was saying so I didn't properly answer some of his questions (I am not a native english speaker, and neither was he).
Interview questions [1]
Question 1
At some point in time there are N cars moving on the same direction on a one-lane road. Their speeds form a random sample of some continuous distribution. Each car will move at its current speed unless it reaches another (slower) car, at which point it will slow down to the speed level of the slower car. What is the expected number of car clusters that will be observed in the long run?
