Jane Street Quantitative Trader Interview Prep: Format, Questions, and a Mock Test
Jane Street publishes more about its trading interview than almost any firm of its size, including a candidate guide that works through the exact concepts its traders ask about. This page pulls that public material into one place: how the loop is structured, what the firm says it is measuring, the myths it goes out of its way to deny, and a knowledge check on the ideas the questions are built from. Everything factual here is sourced to Jane Street's own pages, listed at the end with the date each was read.
AnyLearn is not affiliated with, endorsed by, or sponsored by Jane Street. This page is an independent study aid built from Jane Street's publicly available candidate material.
How the process is structured
| Round | Length | What it tests |
|---|---|---|
| 1.Application[2] | Not published | Jane Street states that it "consider[s] applicants for every open role, not just the one you apply for," so the role you are eventually interviewed against may differ from the one you applied to. |
| 2.Phone interviews[1] | Not published | Foundational trading skills, asked by quantitative traders. Jane Street describes the experience as feeling "a lot more like a conversation than a quiz." No finance knowledge is assumed. |
| 3.Final round, in person[1] | Not published | A variety of question types spanning problem solving, probability and statistics, coding in a language of your choice, data analysis, and general interests. |
Bracketed markers point to the dated sources at the end of this article. Loops change; check the retrieval dates before relying on a round count.
What the firm says it is testing
Jane Street states that its trading interviews are conducted by quantitative traders "asking questions that test a variety of foundational skills that are important for being a successful trader at Jane Street." It lists those skills as "problem solving, probability and statistics, coding (in any language of your choice), data analysis, and general interests."
Two framing statements matter more than the list. The firm says the experience "will feel a lot more like a conversation than a quiz and won't require previous knowledge of finance or involve any complicated math, such as what you might learn in a higher-level math course." And on its general interviewing page it says: "Problem-solving mindset: required. Finance background: optional."
That is an unusual claim, and it is worth taking at face value rather than discounting it. It means preparation aimed at memorising option pricing formulas or macroeconomic theory is aimed at the wrong target. The material the firm actually publishes for candidates is a probability primer.
The four behaviours it names
Jane Street describes successful candidates as "curious, organized, and open-minded" people who "approach interview questions thoughtfully and collaboratively," and it names four specific behaviours:
- Approach the problem methodically.
- Communicate clearly.
- Correct your mistakes.
- Ask why.
Each of those is a process instruction, not a knowledge requirement. "Correct your mistakes" in particular tells you something about how the interview is scored: being wrong and then noticing is a demonstrated skill, not a lost point. That is consistent with the firm's own statement that the interview is a collaboration and that it wants to see how you "take hints and respond to being wrong."
The myths Jane Street explicitly rejects
The firm's candidate guide has a closing section addressing what it calls "common myths" about its interviews. It is the most useful page of the guide, because it contradicts most of the advice circulating about the firm.
| The myth | What Jane Street says |
|---|---|
| You have to be very good at mental math | Numeracy "certainly helps," including for sanity-checking the order of magnitude of an answer, but "we're not going to judge you harshly for not being able to mentally multiply double-digit numbers in 5 seconds." |
| You have to be very good at complicated math | It tries to ask "creative problems that you can reason through with minimal outside knowledge," and "we often prefer the simple, intuitive answer to the one that involves a bunch of esoteric theorems." |
| A wrong answer fails the interview | "We care about how you think about and approach tough problems, which includes how you take hints and respond to being wrong." The reverse also holds: a correct answer alone is not enough, and the interviewer will want the reasoning. |
| You should behave a certain way | "We just want you to be yourself." Saying you are unsure is "totally fine," and taking time to reach a better answer is preferred over answering fast. |
The practical read: the loop optimises for legible reasoning under uncertainty, and speed is not the axis being measured.
Making a market, in the firm's own vocabulary
The one piece of genuine jargon the guide teaches is how to quote a two-sided market. A quote has a direction, a price, and a size. A bid is a commitment to buy, an offer is a commitment to sell.
Jane Street's guide gives the shorthand directly. To buy 10 units at 2 you say "I'm 2 bid for 10." To sell 10 at 4 you say "I have 10 at 4." To do both at once you say "I'm 2 at 4, 10 up." Someone says "sold" to hit your bid, or "take 'em" to lift your offer, and the result is a fill.
The strategy the guide sets out is simple: if you know the expected value, you should be happy to buy below it and sell above it. It then names the three things that set how wide you quote: how much the contract is worth on average, how much you could lose and whether that loss is large relative to your capital, and how much you expect to make given the chance the market actually trades. A market that is too tight makes no money for the risk; one that is too wide never trades.
Key idea: The width of your market is a statement about your own uncertainty, not about the asset.
Adverse selection, the idea the interview circles
The concept that connects the probability questions to the trading questions is adverse selection: the trades you actually get done are worse than they look, because a counterparty chose to do them with you.
Jane Street's guide phrases the working question a market maker should ask as: "Conditional on my order being filled, what do I think the expected value of the security is? Given this updated expected value, would I be happy to have done the trade?" That is conditional probability applied to your own fill. If someone hits your bid immediately, that fill is evidence, and the evidence points toward the true value being below where you bid.
The guide advises being more careful when "the calculations are complicated or the answer is hard to determine," and notes you would "probably be sad to do a trade with a meteorologist about the weather in a week" while being "fine buying a d6 roll for $3." It closes the idea by citing Groucho Marx's line about not wanting to belong to any club that would accept him as a member.
This is why the interview mixes conditional probability with market making. They are the same skill applied to different objects.
What they actually ask
1.Make me a market on the sum of two fair six-sided dice.
What a strong answer coversStart with the expected value: each die averages 3.5, and expectation adds regardless of independence, so the sum is worth 7. Then say something about the spread. A market of 6 at 8 leaves a point of edge on each side; quoting 6.9 at 7.1 is nearly free money for whoever trades with you. Say the number out loud as a market, not as an estimate, because the exercise is testing whether you will commit to a two-sided price. Expect a follow-up that changes the contract, such as taking the maximum instead of the sum, and expect the interviewer to trade against you and ask what you now think.
2.You flip a fair coin until you see two heads in a row. How many flips should you expect to need?
What a strong answer coversThe clean route is recursion rather than summing a series. Define the expected number of further flips from each state: nothing yet, and one head so far. From the start you spend a flip and either land on one head or return to the start. From one head you spend a flip and either finish or return to the start. Writing those two equations and solving gives 6. Strong answers sanity-check the result against the easier case of a single head, which is 2, and note that the general result for a probability p event is 1/p trials to the first success. Setting up the recursion out loud matters more than reaching 6.
3.I roll a fair die and tell you the result is even. What is your expected value now?
What a strong answer coversThe answer is 4, the average of 2, 4 and 6, since conditioning removes the odd outcomes and leaves the remaining three equally likely. The point of the question is not the arithmetic but whether you reach for the conditional structure automatically. Good answers state the general formula, that the probability of A given B is the probability of A and B divided by the probability of B, and note that the information changed the distribution rather than the die. Expect a follow-up that conditions on something less tidy, such as the result being prime, where the answer is the average of 2, 3 and 5.
4.Give me a 90 percent confidence interval for the number of petrol stations in France.
What a strong answer coversDecompose rather than guess: population of roughly 68 million, some sense of how many people one station serves, or a per-town estimate scaled by the number of communes. Say the reasoning aloud, then give an interval wide enough that you would genuinely be surprised to be outside it. Jane Street's own guide warns that people are systematically overconfident, meaning fewer than 90 percent of their 90 percent intervals contain the answer, so a deliberately wider interval is the calibrated response. It also warns against overcorrecting into an interval so wide it says nothing.
5.You quote a market on something you find hard to value, and I hit your bid instantly. What do you now believe?
What a strong answer coversThat the fill is information. Someone chose to sell to you at your bid, which is evidence that the value sits below where you bid, especially when the quantity is hard to evaluate and the counterparty may know more. The strong version restates the guide's own working question: conditional on being filled, what is the expected value now, and would you still want the trade? The follow-up move is to lower your quote rather than repeat it, and to widen when the thing being priced is genuinely uncertain. Saying the fill carries no information is the answer the question is designed to catch.
6.A contract settles at one million times the roll of a fair six-sided die. Where do you buy it?
What a strong answer coversThe expected value is 3.5 million, but expected value alone is the wrong basis for the decision at this size. Buying at 3.4 million and rolling a 1 loses 2.4 million, one time in six. Strong answers say plainly that the size changes the trade: a loss large relative to your capital matters beyond its contribution to the average, because capital lost now is capital unavailable for better opportunities later. Expect the interviewer to ask how your answer changes if you could do the trade a thousand times, which is where the law of large numbers becomes the right tool.
Three sample questions, answered
These three show the level the mock is pitched at, with the answer and the reasoning in the open. The graded paper keeps its answer key server-side.
- 1
- 1.5
- 2
- 4
Why: Set it up recursively. You always spend one flip, and with probability 1/2 you are finished, otherwise you are back where you started. That gives E = 1 + (1/2)E, so E = 2. In general, an event with probability p takes 1/p trials on average to happen once.
- Nothing, because a fill carries no information about value
- You should keep the same quote and wait for a second trade
- The fill is itself evidence that your bid was too high, so the value is probably lower than you thought
- You should raise your bid to trade more of it
Why: This is adverse selection. The counterparty chose to trade against you, so the fact of being filled is information. Jane Street's guide frames the working question as: conditional on my order being filled, what is the expected value now, and would I still be happy to have traded?
- 1/3
- 11/36
- 1/6
- 1/36
Why: Count the complement. Neither roll being a 6 has probability (5/6) squared, which is 25/36, so at least one 6 has probability 1 - 25/36 = 11/36. Adding 1/6 and 1/6 gives 1/3 and double counts the case where both rolls are 6.
An 18-question knowledge check
This is a knowledge check, not a simulation. The real Jane Street loop happens on a whiteboard, in an editor, and in conversation. What this paper does measure is the underlying knowledge those rounds draw on: each question is tagged with a topic, grading happens per topic, and a weak topic points you at the course that fixes it.
- 1.Two fair six-sided dice are rolled. What is the probability that the sum is 10?Counting and combinatorics
- 2.You draw two cards from a standard 52-card deck without replacement. What is the probability that both are clubs?Counting and combinatorics
- 3.Three fair six-sided dice are rolled and the results multiplied. What is the probability the product is odd?Counting and combinatorics
- 4.A contract pays one dollar for every pip showing on a single roll of a fair six-sided die. What is its expected value?Expected value
- 5.Under what condition does E[X + Y] equal E[X] + E[Y]?Expected value
- 6.A six-sided die is weighted so that a 6 comes up half the time, with the other five faces equally likely. What is its expected value?Expected value
- 7.You roll a fair six-sided die and are told the result is at least 4. What is the expected value now?Conditional probability
- 8.You flip three fair coins. Given that at least one is a tail, what is the probability that exactly two are heads?Conditional probability
- 9.Which expression gives the probability of A given B?Conditional probability
- 10.In Jane Street's shorthand, what does "I'm 2 at 4, 10 up" commit you to?Making markets
- 11.You have quoted 3 at 4 on a contract that settles at the value of a die roll. Someone says "take 'em." What is your position?Making markets
- 12.Why would a trader quote a wider market on one contract than another?Making markets
- 13.What does adverse selection mean for someone quoting two-sided markets?Adverse selection
- 14.Jane Street's guide invokes Groucho Marx's line about not wanting to belong to any club that would accept him as a member. Which idea is the comparison making?Adverse selection
- 15.Before quoting a market, which question does Jane Street's guide say you should be asking?Adverse selection
- 16.A contract settles at one million times the roll of a fair die, so its expected value is 3.5 million. Why might a trader still decline to bid 3.4 million?Risk, variance and sizing
- 17.Jane Street's guide warns about a systematic bias in the confidence intervals people give. Which is it?Risk, variance and sizing
- 18.Why does the law of large numbers matter to a trader quoting many small markets?Risk, variance and sizing
Sources
Hiring loops change. Every claim above carries a retrieval date so you can judge how current it is.
- [1]Jane Street, Trading Interviews · retrieved 2026-08-13
- [2]Jane Street, Interviewing · retrieved 2026-08-13
- [3]Jane Street, Probability and Markets (candidate guide, PDF) · retrieved 2026-08-13
- [4]Jane Street, Probability and Markets (guide landing page) · retrieved 2026-08-13
Refresh your memory
Free learning paths covering the ground this loop tests, whatever your score. Each one ends with a shareable certificate.
- BusinessMarket Microstructure: Where a Price Comes From
No exchange quotes a price. It sorts other people's orders and matches them, and every number on a trading screen is the residue of that process. This path builds the limit order book and its matching rules, then answers the question the mechanism leaves open: why competition does not compress the spread to zero. The answer, that a trade is evidence, then explains price impact, the capacity ceiling on any strategy, and why market design decides who profits.
4 lessons - MathProbability and Statistics for Machine Learning
Build the mathematical foundation every ML practitioner needs: go from sample spaces and distributions to Bayesian inference and hypothesis testing. By the end you will be able to choose the right distribution for any modelling problem, derive maximum likelihood estimators, reason about uncertainty the Bayesian way, and correctly interpret p-values and confidence intervals.
4 lessons - BusinessTrading Risk: Measuring It, and Surviving It
Value at Risk compresses a loss distribution into one number, and there is an arithmetic case where it reports that diversification made things worse. This path builds that failure and the coherence axioms explaining it, then attacks the deeper problem: the distribution itself is estimated and wrong in known directions that all understate danger and all fail together. Leverage turns those errors into forced exits. Sizing is the only decision that changes an outcome, and its arithmetic is unforgiving.
4 lessons - BusinessTrade Execution: Working an Order Without Giving It Away
The cost of a trade is not the commission and not the spread. It is the gap between what the decision would have earned on paper and what the account actually got, and the largest component is often the shares that never traded. This path builds implementation shortfall, then the scheduling optimisation that trades impact against timing risk, then the placement decisions underneath every slice. It ends on why execution quality is far easier to produce than to prove.
4 lessons - BusinessOptions and Volatility: Pricing Without Forecasting
An option's value at expiry is arithmetic. Before expiry it looks like it needs a forecast of the underlying, and the central result of the subject is that it does not. This path builds put-call parity, which needs no model at all, then the replication argument that cancels direction and leaves volatility as the only thing being traded. The greeks decompose what a hedged position still carries, and the volatility surface turns out to be the market correcting the model in the model's own units.
4 lessons
Business
Math