About Quantitative Analyst interviews
Quantitative Analyst interviews are among the most technically demanding in finance, blending probability, stochastic calculus, programming, and pricing theory under time pressure. A typical process begins with a recruiter or HR screen confirming credentials (often a PhD or master's in maths, physics, statistics, or financial engineering) and motivation for the desk or asset class. Next comes a hiring-manager call probing your modelling background and which products you've actually touched — rates, credit, equity derivatives, FX, or systematic strategies. The core of the process is a technical loop: expect rapid-fire mental maths, brainteasers, probability puzzles, Ito's lemma derivations, Black-Scholes from first principles, and a coding screen in Python or C++ (often on HackerRank or a live editor). Senior or buy-side roles add a research case study or a take-home where you build, backtest, or critique a model. Final rounds skew toward judgement: how you'd handle model risk, communicate to traders, and defend assumptions to risk committees. Candidates most often stumble by reciting memorised formulae without deriving them, freezing on open-ended estimation, writing fragile code that ignores edge cases, or — critically for senior roles — failing to connect mathematics to P&L and business reality. Interviewers want sharp first-principles thinking, intellectual honesty about model limitations, and the ability to translate quantitative work into decisions traders and risk managers can trust.
Typical stages
- Recruiter screen
- Hiring manager interview
- Technical loop (probability, maths, coding)
- Research case study / take-home
- Final / desk fit
Common formats
- Probability and brainteaser drilling
- Live coding (Python/C++)
- Stochastic calculus and pricing derivations
- Quant research case study
- Behavioral STAR
What hiring managers screen for
- First-principles derivation rather than memorised formulae
- Clean, correct, edge-case-aware code in Python or C++
- Intellectual honesty about model assumptions and limitations
- Ability to connect quantitative output to P&L, risk, and trader decisions
- Comfort with ambiguity in open-ended estimation and research problems
Red flags to avoid
- Reciting Black-Scholes without being able to derive it
- Writing code that ignores numerical stability or edge cases
- Overconfidence in models with no acknowledgement of risk or breakdown conditions
- Inability to explain a model simply to a non-quant stakeholder
- Plagiarised or unexplainable take-home work
Primary questions (14)
Technical
Derive the Black-Scholes PDE from first principles and explain each assumption you rely on.
Why this comes up: It is the canonical test of whether a quant truly understands option pricing rather than memorising the closed-form.
Prep pointers
- Walk through the self-financing replicating portfolio and the delta-hedging argument step by step.
- Be explicit about each assumption: GBM dynamics, constant volatility, no arbitrage, continuous trading, no transaction costs.
- Be ready to state where the no-arbitrage and risk-neutral argument enters, not just the final PDE.
- Common failure: jumping to the final formula — interviewers want the Ito's lemma and hedging logic shown out loud.
Technical
A fair coin is flipped until two consecutive heads appear. What is the expected number of flips?
Why this comes up: Recurrence-style probability puzzles are a staple for testing structured reasoning under pressure.
Prep pointers
- Set up states (no progress, one head) and write expectation equations for each.
- Talk through your reasoning aloud — interviewers score the method as much as the answer.
- Sanity-check the final number against intuition before committing.
- Common failure: guessing or brute-forcing instead of building the Markov-state recurrence.
Technical
How would you calibrate a volatility surface, and how do you handle arbitrage-free constraints?
Why this comes up: Calibration is core day-to-day quant work and reveals practical modelling depth beyond textbook theory.
Prep pointers
- Distinguish between interpolation/parametric approaches (e.g. SVI) and full local/stochastic vol models.
- Explain calendar and butterfly arbitrage conditions and how you enforce them.
- Mention how you balance fit quality against smoothness and stability of greeks.
- Common failure: describing a fit procedure with no mention of arbitrage-free or numerical robustness considerations.
Technical
Write a function to price a European option via Monte Carlo and discuss how you'd reduce variance.
Why this comes up: It tests coding competence alongside numerical methods knowledge in one combined task.
Prep pointers
- Structure clean, vectorised code and handle the discounting and payoff clearly.
- Discuss variance reduction: antithetic variates, control variates, quasi-random sequences.
- Mention convergence rate (1/sqrt(N)) and how you'd validate against the analytic price.
- Common failure: producing slow loop-based code or ignoring how to verify correctness.
Behavioural
Tell me about a time a model you built behaved unexpectedly in production or testing.
Why this comes up: Model risk is central to the role, and hiring managers want to see how you detect and respond to failure.
Prep pointers
- Pick an example where you caught the problem yourself rather than someone else flagging it.
- STAR: Situation = the model and its use; Task = what broke or diverged; Action = how you diagnosed and fixed it; Result = quantified impact and what changed in your process.
- Emphasise the diagnostic process and the controls you added afterwards.
- Common failure: framing it as someone else's fault or omitting what you learned.
Behavioural
Describe a time you had to explain a complex quantitative result to a non-technical stakeholder.
Why this comes up: Quants who cannot translate maths into business language rarely get their work used.
Prep pointers
- Choose a trader, risk manager, or PM as the audience to make it concrete.
- STAR: Action should show how you stripped jargon and anchored on the decision they needed to make.
- Highlight the outcome — did your explanation change a decision or build trust?
- Common failure: describing the maths in detail instead of the communication technique.
Behavioural
Tell me about a research idea you pursued that ultimately didn't work.
Why this comes up: It probes intellectual honesty and whether you can kill your own ideas with evidence.
Prep pointers
- Pick something genuinely promising that you abandoned for sound reasons.
- STAR: Action should cover how you tested it and what evidence convinced you to stop.
- Show the discipline of not overfitting or rationalising a failing signal.
- Common failure: pretending everything you've tried succeeded, which reads as dishonest.
Behavioural
Describe a time you disagreed with a trader or senior researcher about a modelling choice.
Why this comes up: Quants must defend rigour while respecting desk experience and P&L pressures.
Prep pointers
- Choose a disagreement resolved through data or analysis, not personality.
- STAR: Action should show how you presented evidence while staying open to their market intuition.
- Note how the resolution affected the model or process.
- Common failure: portraying yourself as always right or as a pushover.
Situational
A pricing model is producing P&L that diverges from the desk's expectation right before close. What do you do?
Why this comes up: It tests judgement, prioritisation, and risk awareness under realistic time pressure.
Prep pointers
- Lead with triage: isolate whether it's data, market move, or model error.
- Discuss interim controls — flagging risk, escalating, using a conservative fallback.
- Show you'd communicate uncertainty clearly rather than hide it.
- Common failure: diving straight into the maths without managing the live risk first.
Situational
You're handed a legacy model with no documentation that the desk relies on daily. How do you approach it?
Why this comes up: Inheriting opaque models is common, and it reveals systematic, risk-conscious working habits.
Prep pointers
- Describe reconstructing assumptions through testing and reverse-engineering outputs.
- Mention building regression tests before changing anything.
- Show how you'd document and identify hidden risks without disrupting users.
- Common failure: proposing a full rewrite immediately without understanding why it works.
Competency
How do you decide whether a backtested strategy or signal is robust rather than overfit?
Why this comes up: Distinguishing genuine edge from noise is the defining competency of a quant researcher.
Prep pointers
- Discuss out-of-sample testing, walk-forward analysis, and parameter sensitivity.
- Mention transaction costs, capacity, and regime dependence as reality checks.
- Reference multiple-testing / data-mining bias and how you guard against it.
- Common failure: relying on a single impressive backtest curve as proof.
Competency
Walk me through how you validate a model before it goes into production.
Why this comes up: Model validation discipline separates production-ready quants from academic ones.
Prep pointers
- Cover benchmarking against known cases, sensitivity analysis, and stress scenarios.
- Mention independent review, documentation, and sign-off processes.
- Show awareness of regulatory or governance expectations where relevant.
- Common failure: equating validation with simply matching one closed-form result.
Competency
How do you choose between model complexity and parsimony when accuracy and explainability conflict?
Why this comes up: It reveals mature judgement about the trade-offs that govern real quant work.
Prep pointers
- Frame the decision around the use case: pricing, risk, or signal generation.
- Discuss stability, interpretability, and cost of failure as deciding factors.
- Give a concrete example where you deliberately chose the simpler model.
- Common failure: defaulting to 'more complex is better' with no trade-off reasoning.
Culture fit
Why this asset class or desk, and how do you stay current with quantitative research?
Why this comes up: Desks want genuine interest and self-driven learning, not a generic quant looking for any seat.
Prep pointers
- Be specific about why this product or strategy interests you intellectually.
- Reference concrete papers, sources, or techniques you follow.
- Connect your background to the desk's actual problems.
- Common failure: a generic answer that would fit any quant role equally well.
More practice questions (13)
Technical
State Ito's lemma and apply it to derive the dynamics of log(S) under geometric Brownian motion.
Why this comes up: It's a fundamental tool every derivatives quant must wield fluently.
Technical
What is the difference between historical, implied, and realised volatility, and when does each matter?
Why this comes up: Volatility concepts are central and frequently confused by weaker candidates.
Technical
How would you estimate the value of pi using a Monte Carlo simulation?
Why this comes up: A classic warm-up testing simulation intuition and basic numerical reasoning.
Technical
Explain the difference between risk-neutral and real-world probability measures.
Why this comes up: Misunderstanding measures is a common gap that interviewers probe directly.
Technical
Given a matrix, how would you check it's a valid covariance matrix and fix it if it isn't?
Why this comes up: Positive semi-definiteness issues arise constantly in portfolio and risk work.
Technical
What numerical methods would you use to solve a PDE for an American option, and why?
Why this comes up: It tests practical knowledge of finite difference methods and early-exercise handling.
Technical
How do you simulate correlated random variables, and what is the Cholesky decomposition's role?
Why this comes up: Correlated simulation underpins most multi-asset pricing and risk engines.
Situational
A data vendor feed has gaps and occasional spikes. How do you handle it before modelling?
Why this comes up: Data hygiene is a daily reality and poor handling silently corrupts models.
Situational
Your model's runtime is too slow for the desk's needs. How do you approach optimisation?
Why this comes up: Performance trade-offs are common where pricing must keep pace with markets.
Competency
How do you document and version-control your models and research code?
Why this comes up: Reproducibility and governance are increasingly scrutinised in quant teams.
Competency
How do you prioritise when multiple traders want competing model improvements at once?
Why this comes up: It reveals stakeholder management and judgement about business value.
Behavioural
Tell me about a time you automated or streamlined a repetitive quantitative process.
Why this comes up: It shows initiative and the engineering mindset desks increasingly expect.
Culture fit
Would you rather work on a fast-moving trading desk or a longer-horizon research function, and why?
Why this comes up: It checks fit with the team's actual pace and working style.
Researching the Quantitative Analyst role?
See the full skills, salary and market breakdown — what employers
actually want, and the biggest skills gaps.
Quantitative Analyst skills & salary →
Get a prep pack tailored to your experience
describe.me matches these questions against your real work history,
flags your prep priorities, and gives you a STAR scaffold per question.
Start free →