How to Prepare for a Quant Trading Interview: The Complete Guide

The Hiring Pipeline: From Screen to Offer

Almost every quant trading firm — Optiver, IMC, Jane Street, Flow Traders, SIG (Susquehanna International Group), Akuna, CTC (Chicago Trading Company), Maven, and market-making businesses such as Citadel Securities — runs some version of the same funnel: an online screen, one or two interview rounds, then a concentrated final round often called a superday.

Stage one is a timed online assessment, typically run through a third-party testing platform or a firm-built tool. It usually bundles a mental math speed test with a numerical or logical reasoning section, and increasingly a short gamified module meant to score traits like decision-making under uncertainty rather than a single "right answer." Pass this and you'll typically get a 30-60 minute phone or video interview: a handful of probability or expected-value problems, sometimes a quick math re-check, and a couple of fit questions.

The superday compresses everything into one day: several back-to-back interviews, typically in the 30-45 minute range, covering probability and brain teasers, a trading or market-making game, technical or coding rounds (heavier for research and dev-adjacent roles), and behavioral fit — sometimes blended into the same session rather than separated. Exact round count, timing, and structure vary by firm and by year, so treat this as a typical shape rather than a fixed script. The full pipeline commonly runs several weeks from first assessment to offer, though it compresses fast during graduate recruiting cycles.

Inside the Mental Math Speed Test

The format popularized by Optiver's "80-in-8" — 80 arithmetic questions in 8 minutes, no calculator — is the most publicly documented version of this test and the one most candidates benchmark against first. Other firms, including Flow Traders, SIG, Akuna, Maven, and CTC, run their own timed numerical tests that vary in question count, length, and format from firm to firm and year to year, sometimes splitting arithmetic from a separate pattern-recognition or sequence section.

The content is deliberately unglamorous: two- and three-digit multiplication and division, percentage changes, fraction-to-decimal conversions, quick compounding, and converting between odds and implied probability. There's no calculus and no advanced formulas — the test measures fluent, low-error arithmetic under time pressure, because that's what adjusting quotes on a fast-moving order book actually requires.

Because there's no partial credit for showing work, the skill being tested is retrieval speed, not derivation. The highest-leverage prep isn't learning new math — it's converting things you already technically know (17% of 640, 13 × 24, 5/8 as a decimal) from a "figure it out" process into an instant, automatic answer. Firms use these scores as a cheap, mechanical filter early in the funnel, before investing any interviewer time.

Why Negative Marking Changes Your Strategy

Most mental math speed tests penalize wrong answers — commonly deducting more than a correct answer is worth (for example, losing two points for a miss against gaining one for a hit) — while a skipped question costs nothing. That asymmetry means the test isn't scored like a typical school exam, and treating it that way is the single most common strategic mistake candidates make.

Run the expected-value math: if a wrong answer costs twice what a right answer earns, a pure guess needs to be right at least two-thirds of the time (about 67%) just to break even. Skipping is roughly EV-neutral; guessing without any basis is usually EV-negative. The rational approach is triage, not speed-maximizing: for every question, decide within a couple of seconds whether it's "solve now," "skip," or "worth eliminating options and taking a calculated risk on." Attempting 70 questions at 90% accuracy beats attempting all 80 at 75% under most of these scoring schemes.

Practically, this means training two separate skills — raw calculation speed and a fast, honest self-assessment of confidence — rather than just drilling to answer faster. Candidates who only practice speed tend to see accuracy collapse exactly when the clock is real.

Brain Teasers and Probability Rounds

Once you're past the numerical screen, interviews shift from arithmetic to reasoning. Expect expected-value problems (pricing a bet, valuing a game with uneven payouts), conditional probability and Bayesian updating, combinatorics, and Fermi-style estimation. Classic logic riddles still appear but are a smaller share of what's asked than pure probability and EV reasoning.

What's graded is rarely just the final number. Interviewers often already know the answer and are watching how you get there: whether you state assumptions out loud, define variables before computing, break a messy problem into smaller solvable pieces, and sanity-check the magnitude of your result. Many will deliberately push back on a correct answer or add a twist mid-problem — not because you're wrong, but to see whether you can adapt your reasoning live rather than freezing or over-defending a memorized approach.

The practical implication: memorizing answers to famous brain teasers is close to useless, because the interviewer will vary the setup. Practicing the underlying toolkit — expected value, Bayes' rule, combinatorics, structured estimation — transfers to whatever specific puzzle actually shows up.

Trading Games and Market-Making Simulations

Many superdays include a live trading or market-making game, sometimes one-on-one with an interviewer, sometimes group-based. You're typically given partial information — your own cards, a running news feed, or asymmetric signals — and have to quote a bid/ask spread, trade, and manage a running position as new information arrives.

What's assessed isn't whether you "win" the game. It's whether your spread width reflects your actual uncertainty (wide when unsure, tight when confident), whether you update your fair-value estimate correctly as information arrives, whether you track and manage position size and risk rather than trading recklessly, and how you handle being picked off by a better-informed counterparty — an unavoidable part of real market making. Composure after a bad trade is scored as much as the math.

You can't fully simulate the social and competitive pressure of a live multiplayer round through solo practice, but you can prepare the underlying instincts: drill quick expected-value recalculation, understand how spread width should scale with uncertainty and inventory, and learn the basic economics of market making so the concepts aren't new when the game starts.

Behavioral and Fit Interviews

"Why trading?" and "why this firm?" show up in nearly every process, and treating them as a formality is a mistake — interviewers use them to filter for people who understand what the job is actually like day to day: long stretches with no clear feedback, real money lost on well-reasoned decisions, and a competitive, sometimes repetitive grind that "the intellectual challenge" alone doesn't fully capture.

Prepare specific stories, not talking points: a time you made a decision with incomplete information, a mistake you owned and what changed afterward, a moment you took a calculated risk that didn't pay off, and times you took feedback well or handled disagreement. Structure them tightly — situation, action, outcome, and what it says about how you think — rather than narrating a resume line item.

Fit is also assessed continuously, not just in a dedicated round — many technical interviewers are quietly judging composure, coachability, and how you take a hint or pushback on a brain teaser as part of the same conversation. Being defensive when corrected, or visibly rattled after getting a probability question wrong, is remembered as much as the math itself.

Firm-by-Firm Notes

Processes change often and firms rarely publish exact formats, so treat this as general orientation based on public candidate reports, not a guaranteed script — always check the firm's current careers page and recent candidate reports before an interview.

Optiver is the firm most publicly associated with a branded mental math speed test and is widely reported to pair it with a strong focus on options and market-making intuition in later rounds. Jane Street is well known for unusually deep probability and expected-value puzzles combined with trading games and, for some roles, technical coding rounds. Flow Traders, IMC, and CTC are generally described by candidates as leaning heavily on speed-math testing and trading simulations. SIG's process is commonly reported to combine probability testing with a structured behavioral or psychometric component. Some firms, including Akuna and Maven Securities, are reported to use third-party numerical or gamified assessment tools early in the funnel. Citadel Securities (the market-making business) and Citadel's quantitative investing business are separate operations from the same parent group; both, along with other large multi-strategy and quant funds, tend to weight coding ability and applied statistics more heavily relative to pure mental-arithmetic speed — especially for research-track roles.

The common thread: every one of these firms tests fast, accurate quantitative reasoning under pressure — they just weight the ingredients differently.

Common Mistakes That Sink Otherwise-Strong Candidates

Training pure speed before accuracy is stable: rushing from day one builds bad habits that collapse under real time pressure and negative marking. Fix the error rate first, then compress the time.

Treating brain teasers as a bank to memorize: interviewers vary numbers and add twists specifically to catch memorized answers; practice the underlying method, not a library of exact solutions.

Never practicing under real time pressure: solving problems untimed at home doesn't transfer to an 8-minute countdown with a proctor watching. Simulate the actual conditions — same time limit, no calculator, no pausing — well before test day.

Skipping behavioral prep because "it's just the math firm": fit and composure are graded even inside technical rounds, and a candidate who nails every brain teaser but seems rattled or evasive on "why trading" is a common reason for a post-superday rejection.

Preparing for only one firm's known format: firms share more DNA than differences, but expecting exactly one test style when a firm has changed its assessment provider is a frequent, avoidable surprise. Prepare the underlying skills, not a single firm's exact test.

How long should I spend preparing for a quant trading interview?

Most candidates who pass benchmark four to six weeks of consistent daily practice (20-40 minutes) rather than a short cram, because mental math fluency and calibrated brain-teaser reasoning are built through repetition, not crammed the night before. If you've already built the base skills from a prior cycle, one to two weeks of refresh practice is often enough.

Do I need a finance or economics degree to get hired?

No. Most of these firms hire heavily from math, physics, engineering, computer science, and other quantitative degrees, and generally do not require finance background — they test raw quantitative reasoning and teach trading-specific knowledge on the job. What matters is comfort with probability, fast arithmetic, and structured problem-solving, not prior markets experience.

What counts as a good score on a mental math speed test like the 80-in-8?

Firms don't publish official cutoffs, but candidates commonly report needing to answer roughly 55-65+ of 80 questions correctly, once negative marking is factored in, to reliably clear the bar. Treat any specific number you see online as an anecdotal data point, not a guarantee, and aim comfortably above whatever threshold you've seen reported rather than right at it — nerves alone can cost several questions on the real day.

Should I guess when I'm not sure of an answer?

Only if you've eliminated enough options that your remaining odds beat the test's break-even threshold — usually around two-thirds (roughly 67%) once you factor in a typical two-for-one negative-marking penalty. An unfounded guess is usually worse than skipping, so build the habit of a fast, honest confidence check before answering rather than guessing by default.

How important is coding for these interviews?

It depends heavily on the role and firm. Pure trading roles at prop shops like Optiver, Flow Traders, and IMC tend to weight mental math, probability, and market intuition most heavily and may ask little or no coding, while quant research roles, quant developer roles, and some trading roles at firms like Jane Street or Citadel Securities test coding and applied statistics more heavily. Check the specific role description rather than assuming.

What's the difference between a "trader" and "quant researcher" interview track?

Trader tracks generally emphasize speed, probability intuition, market-making games, and composure under live pressure. Quant researcher tracks generally emphasize deeper statistical and mathematical rigor, sometimes coding, and longer-form problem-solving over live speed. Many firms share an initial numerical/probability screen across both tracks before the paths diverge.

Can a practice app actually replicate the real test conditions?

A well-built timed drill app can closely replicate the mechanical conditions of the mental-math screen — question format, timing, and negative marking — which is genuinely useful for building speed and calibration under pressure. It can't replicate the live, adaptive back-and-forth of a brain-teaser or market-making interview, which still needs practice talking through problems out loud with a real person or study partner.