Optiver Data Scientist Interview Questions

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Optiver data scientist interview with a pretty gnarly market-making game problem spanning five sub-parts. The whole thing was essentially one long quantitative case study dressed up as an interval estimation exercise. Left feeling like I'd passed a math exam I hadn't studied for.

Given a continuous probability distribution over an unknown quantity, how would you set up the optimization problem to choose interval endpoints that maximize expected score in the market-making interval game?
Derive the first-order conditions for the optimal interval and explain how the optimal coverage probability differs between light-tailed and heavy-tailed belief distributions.
How would you allocate risk across five rounds to hit a total score of at least 2.0? Walk through a dynamic programming or heuristic approach and explain how you'd adjust interval tightness based on your running score.
Under a log-normal belief about the true answer, describe a quick mental method for picking near-optimal interval endpoints. How would you adapt this for heavy-tailed beliefs?
How would you design a simulation to validate your five-round interval strategy and estimate the probability of hitting the 2.0 score target?

“This took me a minute to even parse.”

View Post

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Optiver Data Scientist interview with a heavy math/combinatorics problem involving Latin squares. The question was proof-based, which I did not expect for a DS role, and there were follow-ups that pushed into symmetry arguments and edge cases.

You have an n×n Latin square (every row and column is a permutation of 1 through n). Let L be the sum of all entries on or below the main diagonal. Prove that L is at least n(n+1)(n+2)/6, and show a construction that achieves equality.
What is the tight lower bound for the sum strictly below the main diagonal (excluding diagonal cells), and does your equality construction still work?
What is the maximum possible value of L, and how does it relate to the minimum via some symmetry of Latin squares?
Your proof of the lower bound probably only used one of the two permutation constraints (rows or columns). Which one, and does the tightness construction still require both?
How does the argument change if you consider the region on or below the main anti-diagonal instead of the main diagonal?

“I stared at this for a while before anything clicked.”

View Post

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Optiver Data Scientist interview with a classic probability/random walk puzzle. The kind of question where you either see the structure quickly or spend ten minutes drawing octagons on a whiteboard.

Two particles start at opposite vertices of a regular octagon, four steps apart. Each time step, each particle independently flips a fair coin and moves one step clockwise or counterclockwise. What is the expected number of total coin flips until both particles land on the same vertex?

“The key insight I kept fumbling around before landing on: you can reduce this to tracking the relative gap between the two particles, which itself does a random walk on a cycle.”

View Post

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Optiver Data Scientist interview with a math-heavy algorithmic problem that looks deceptively like a simulation exercise but punishes you hard if you actually try to simulate it. The whole thing hinges on recognizing a number theory pattern fast enough to implement something logarithmic.

You have two non-empty piles of stones with sizes a and b. Each move, you double the smaller pile by transferring that many stones from the larger one. Determine whether this process will eventually empty one of the piles, or cycle forever. Your solution must run in logarithmic time.

“My first instinct was just to simulate it and slap a visited-states check on top.”

View Post

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Interviewed for a Data Scientist role at Optiver and got hit with a pretty deep sequence inference problem that was more of an algorithms design question than anything I expected. The whole thing felt like a quant-flavored coding challenge dressed up as a data science interview.

Given an integer sequence of up to 50 terms that may follow arithmetic, geometric, alternating, polynomial, digit-level, or interleaved patterns, describe a systematic approach to infer the governing rule and predict the next several terms. Cover the heuristics you'd apply in order, how you'd handle ambiguous or noisy sequences, and how you'd implement this as a scored, ranked program with time complexity analysis.

“This one took me a while to even parse.”

View Post

Optiver·Data Scientist·Recruiter / HR Screen

Jun 2026

Standard HR screen for a Data Scientist role at Optiver. Nothing too surprising, mostly behavioral with a market-making motivation angle thrown in. The questions were predictable but Optiver's context makes some of them feel a bit more loaded than usual.

Walk me through your background and what you've been working on.
Why do you want to work at Optiver, and what draws you to market making specifically?
What makes you a strong fit for this role specifically?
Tell me about a time you had to make a fast, high-stakes decision with incomplete information. What happened?
What are your biggest strengths, and what's a weakness you're actively working on?
Describe a conflict you had with a teammate and how you worked through it.
Tell me about a failure, what you took away from it, and how it changed how you work.
Do you have any questions for us?

“I always fumble the opener a little.”

View Post

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Optiver Data Scientist interview with a probability/statistics problem that looks deceptively clean but has a real bite to it once you start working through the gap distribution. The follow-ups were no joke either.

Buses on a route are scheduled every x minutes. Each bus independently arrives on time with probability 1/2, or exactly x/2 minutes late with probability 1/2. You arrive at the stop at a uniformly random time. What is your expected waiting time in terms of x?
How does your answer change if each bus's delay is uniform on [0, x/2] instead of the binary on-time or x/2-late setup?
Generalize: if a bus is late by d minutes with probability p and on time otherwise (with d < x), what is the expected wait as a function of x, p, and d?
Using your gap-based formula, explain intuitively why irregular headways always increase a random rider's expected wait, even when the mean headway stays the same.
How would the expected wait change if buses arrived as a Poisson process with mean headway x instead of the scheduled setup?

“I went straight for the naive E[G]/2 and felt pretty good about it for a moment.”

View Post

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Optiver data science interview had a timed probability question that was genuinely stressful. The 90-second clock made it hard to think clearly, and the extensions they asked for afterward were where things got interesting.

An urn has r red, b blue, and g green balls. You draw 3 without replacement. What's the probability of getting at least 2 reds? You have 90 seconds, and they also want a mental-math shortcut that generalizes.

“The 90-second cap is real and they mean it.”

View Post

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Optiver data scientist interview hit me with a pretty deep algorithmic puzzle that I wasn't expecting to go so far into the math. The whole session was basically one extended problem with a lot of follow-up layers.

You have an m×n grid of lights where toggling a cell also flips its orthogonal neighbors. Given a starting configuration and a target, find the minimum number of toggles to reach the target, or determine that it's impossible.

“This wrecked me a little because I started thinking BFS and they clearly wanted something more principled.”

View Post

Optiver·Data Scientist·Technical Phone Screen

Jun 2026

Optiver Data Scientist interview had a math/logic puzzle that looked deceptively simple on the surface. One coding question about pile balancing, and it required actually thinking through the structure of the problem rather than just throwing a loop at it.

You have two piles of apples with sizes a and b. In each move, you pick one pile as the receiver and double it by moving apples from the other pile (only legal if the giving pile has enough). Can you reach a state where both piles are equal? Return true or false.

“Spent the first few minutes trying to simulate it with a BFS and realized that could blow up badly given the constraint range goes up to 10^9.”

View Post