Market making interview questions, worked through
A market-making question asks you to quote a bid and an ask on an uncertain quantity, then react as the interviewer trades with you. Center the quote on your best estimate, set the width from your uncertainty, and treat every trade as information: if someone buys from you, they may know something, so move your quote up. Say each of these steps out loud; the reasoning is what is scored.
Key points
- Quote around your fair value; the width reflects how unsure you are and how informed the other side may be.
- A trade against you is information. Update the fair value before quoting again.
- Inventory matters: after selling, skew your quote up to attract sellers and discourage buyers.
- Say the fair value, the width and the reason for each change in one sentence.
- Keep the arithmetic simple and round on purpose; speed is part of the test.
The terms, in one table
| Term | Meaning | Example |
|---|---|---|
| Bid | The price you will buy at | 9.5 |
| Ask (offer) | The price you will sell at | 11.5 |
| Width | Ask minus bid | 2 |
| Fair value | Your best estimate of the quantity | 10.5 |
| Skew | Shifting both prices to manage inventory or new information | Moving to 10 at 12 after selling |
A worked dialogue: three dice
The quantity is the sum of three fair six-sided dice, not yet rolled. Each die averages 3.5, so the sum averages 10.5. The variance of one die is 35/12, so the sum has variance 8.75 and a standard deviation of about 3.
With no one holding private information, a tight quote around 10.5 is fair, for example 9.5 at 11.5. The interviewer buys one at 11.5. Since nobody has seen the dice, that trade tells you nothing about the sum; your fair value stays 10.5, and you might skew slightly upward because you are now short one.
Twist: the interviewer says they have seen one of the three dice, then buys again at 11.5. If the die they saw shows x, their fair value is x + 7. Buying at 11.5 makes sense only if x + 7 > 11.5, so x is 5 or 6. Given that, the expected sum is 5.5 + 7 = 12.5. Move the quote up, skew it a little higher because you are now short after two sales, and widen it while you are less sure who knows what: 12 at 14 is a reasonable answer.
How it sounds out loud
“Fair value is 10.5: three dice at 3.5 each. Standard deviation is about 3, so I’ll make it 9.5 at 11.5.”
“You bought at 11.5. Nobody has seen the dice, so I don’t learn anything about the sum. I’m short one, so I’ll lean up a little: 9.75 at 11.75.”
“If you have seen a die and still buy at 11.5, that die is a 5 or a 6. My fair value moves to 12.5, and I’m short, so I’ll lean up: 12 at 14 until I learn more.”
How wide should the market be?
Width trades two risks. Too tight and an informed counterparty picks you off; too wide and nobody trades, which in an interview reads as avoiding the question. A useful rule is to start with a width of about one standard deviation of your estimate when the other side might be informed, and tighter when they clearly are not. Say the rule you use; interviewers mostly want to hear that width is a decision with a reason.
For quantities you have to estimate rather than compute, such as the number of floors in a building you can see, the uncertainty is in your knowledge, so the width is wider. State a range you would be 80% sure of, then quote inside it.
A market on something you have to estimate
Not every market is on dice. Suppose the question is the total weight, in kilograms, of 1,000 one-euro coins. You probably do not know the weight of one coin, so reason from what you do know: it feels heavier than a paper clip and lighter than a battery, so somewhere between 5 and 10 grams. That makes the total 5 to 10 kilograms, with a central guess of 7.5.
Quote inside your range, for example 6.5 at 8.5, and say that the width comes from your uncertainty about one coin, multiplied by 1,000. If the interviewer then sells to you at 6.5 several times, ask yourself whether they know the answer. If they might, lower your fair value and widen; interviewers also trade against you to test the update, whatever the truth is. The real figure is 7.5 grams per coin, so 7.5 kilograms. The score depends on the reasoning and the updates; nobody expects you to know a coin’s weight.
Follow-ups to expect
- What is your position now? Count the trades: after selling twice you are short two.
- What is your expected profit on that trade? Selling at 11.5 when fair value is 10.5 earns 1 per unit on average, before you learn anything.
- Would you trade larger size? Larger trades from an informed counterparty cost more, so widen or quote a smaller size.
- Where would you quote if I told you the first die is a 2? The fair value becomes 2 + 7 = 9, so recentre around 9.
Common mistakes
- Quoting a single number instead of a bid and an ask.
- Not updating after a trade, or updating the same way whether or not the counterparty could know more.
- Making the market so wide it is useless, to avoid being picked off.
- Long silent arithmetic instead of a rounded estimate and a quick quote.
- Forgetting inventory: after selling several times, still quoting symmetrically around the old value.
Questions
Do quant developer interviews ask market making?
Mostly in trader and hybrid roles; some developer loops include a short version. Practise it if your target role is close to trading.
Is there a single correct quote?
No. Interviewers look for a fair value you can justify, a width with a reason, and sensible updates. Two candidates can quote differently and both do well.
How fast should I answer?
A first quote within a few seconds is typical. Round the arithmetic and refine after the first trade rather than computing everything exactly first.
Practise this out loud with Vera
Mockich’s probability round runs 30 min: five short problems on expected value, combinatorics, conditional probability, estimation and market making, reasoned out loud. Once you answer, Vera asks you to sanity-check it or extends the problem once, and gives at most two small nudges. Market-making questions are asked like the others; there is no live trading game. The report scores correctness, structured reasoning, speed and estimation, communication (plus market intuition at lead) and quotes what you said.
Related guides
Sources
We read these sources on 2026-11-03 for how market-making questions are run and scored; the dice dialogue, its numbers and the width rule of thumb are our own framing.
- [1] dwcoder on GitHub, An Interview Primer for Quantitative Finance (CC BY-SA 4.0)
- [2] mikinty on GitHub, Quant Trading Interview Questions (games and market questions, topic reference only)
- [3] techinterview.org, What is a quant trading interview?
- [4] TraderMath, Quant interview questions