How Hudson River Trading's process runs, as candidates and the firm describe it, and the questions candidates report, each with a worked solution. Free, no sign-up.
Quantitative trader candidates report probability questions such as conditional expectations with dice.
Algorithm trader candidates report a coding test; one was asked to find the differences between two data files. A 2026 report describes a deep dive into a past research project and a design-and-code task (a cache for model checkpoints).
Try each one before opening the solution. Interviewers change the numbers, so learn the method.
Quantitative Trader · Interview · Conditional probability · reported by a candidate
You roll a fair die until the first 6. Given that a 3 appeared before that first 6, what is the expected number of rolls (including the 6)?
Expected number of rolls, given a 3 came first:
Answer and worked solution
Answer: 9
Key idea. Split the game at the first roll that is a 3 or a 6. When it happens doesn't depend on which of the two it is.
Each roll is a 3 or a 6 with probability 1/3, so the first such roll comes after 3 rolls on average. It's a 3 or a 6 with probability ½ each, independently of when it happens.
Given that it was a 3, no 6 has appeared yet, and you start a fresh wait for a 6: 6 more rolls on average.
E[rolls | a 3 came first] = 3 + 6 = 9.
Check. Unconditionally it's 6. Knowing a 3 came first means the game didn't end at that point, so it has to be longer.
Trap. Answering 6, or arguing that the 3 "doesn't matter". Or summing a series: it works (it gives 9) but takes far longer.
Interviewers may push further: Compare: expected rolls until a 6, given every roll before it was even. (1.5, not 3.)
Trader · Interview · Behavioural · reported by a candidate
Behavioural questions candidates report:
Explain one of your research projects in detail. What exactly was your contribution?
Answer one out loud, as you would in the interview.
Model answer
Use a short story: the situation in one sentence, what you did (not the team), the result with a number if you have one, and what you'd do differently. Aim for about two minutes.
Expect to be pushed on details: why that method, what you tried that failed, how you'd know if the result was wrong. Pick a project you can defend line by line.
Algo Developer (campus) · Interview · Expected value · reported by a candidate
A point is placed uniformly at random in a 1 × 1 square.
Reported as "expected distance to an edge, with a randomly placed object". We use the usual version: a point uniform in a unit square, distance to the nearest edge.
What is the expected distance to the nearest edge?
Answer and worked solution
Answer: 1/6 ≈ 0.16667
Key idea. Use the tail formula E[D] = ∫ P(D > t) dt.
D > t means the point is at least t from every edge: inside a central square of side 1 − 2t.
P(D > t) = (1 − 2t)² for 0 ≤ t ≤ ½.
E[D] = ∫₀½ (1 − 2t)² dt = 1/6.
1/6.
Check. The farthest any point can be from an edge is ½ (the centre), and most points are near an edge, so a small answer makes sense.
Trap. Answering ¼ (the mean distance to one particular edge, not the nearest).