Hedge Funds interview preparation

Questions

100

Traced to a firm

39

Firms

16

Updated

September 2026

Asked at

All firmsMan Group10Balyasny Asset Management7Bridgewater Associates3DED.E. Shaw3Apollo Global Management2KKR2Oaktree Capital Management2Point722SCSquarepoint Capital2ACAQR Capital Management1BGBaupost Group1Coatue Management1HPS Investment Partners1Northern Trust1Viking Global Investors1Wolverine Trading1

Topic

All topicsStrategy taxonomy8Stock pitch10Short selling6Portfolio construction8Risk and drawdown8Performance and alpha7Event-driven and merger arb8Distressed and credit5Fund structure and economics7Financing, NAV and operations6Compliance and research process5Quant and systematic6India and Category III AIFs5Career and fit11

Level

AnyCoreIntermediateHard

Type

AnyTechnicalMarket viewBrainteaserCaseFit

1. 075What is material non-public information, and where exactly is the line?Compliance and research processHardtechnicalComplianceMulti-manager platforms

   Say this

   Material means a reasonable investor would consider it important in deciding whether to trade, which in practice means it would move the price. Non-public means it has not been broadly disseminated. Both tests have to be met, and the difficulty in real life is almost never the definition, it is the mosaic question.

   Then walk it

   1. Materiality examples that are clearly over the line: unreported results or guidance, an unannounced deal, a pending regulatory decision, a major customer loss, a CEO departure, an unannounced buyback.
   2. Non-public means not broadly disseminated. A fact told to one hedge fund, or sitting in a document that was not distributed, is non-public even if it was not marked confidential. Being told it by accident does not make it public.
   3. The legitimate discipline is the mosaic theory: assembling many individually non-material, public or lawfully obtained pieces into a conclusion nobody else has. Channel checks, satellite imagery, credit card panels, job postings, pricing scrapes. That is the entire alternative data industry and it is legal precisely because no single piece is material and non-public.
   4. Where the line actually blurs: a supplier telling you their shipments to a customer are down sharply. Non-public, arguably material, and the supplier may have a duty. The test involves how you got it and whether anyone breached a duty in passing it on, which is the misappropriation and tipping analysis.
   5. So the practical rules are procedural, not intellectual. Restricted and watch lists, pre-clearance of personal trades, chaperoned expert calls, information barriers between pods at platforms, and a habit of escalating anything ambiguous rather than resolving it yourself.
   6. The honest thing to say: the rule is asymmetric on purpose. The cost of escalating something harmless is an hour of compliance time; the cost of being wrong is criminal. So I would rather be the analyst compliance hears from too often than the one they hear about from a regulator.

   Where candidates lose it

   Trying to look sophisticated by arguing about grey areas. Interviewers are checking your instinct, and the correct instinct is to escalate rather than adjudicate. Do mention the mosaic theory, because it shows you know where the legitimate edge lives, but pair it with the procedural controls.

   Expect next

   • Is a sell-side analyst's unpublished view MNPI?
   • How do information barriers work between pods?
   • Where does alternative data cross the line?
2. 076How can you make a financial model detailed enough to be useful but simple enough that you can cover a lot of companies?Compliance and research processHardtechnicalBalyasny Asset ManagementEquity Research · New York · 2026

   Say this

   Model the two or three drivers that actually move the stock in detail and leave everything else as a ratio. The rule I use is that a line gets its own build only if a reasonable disagreement about it changes my target price by more than a few percent. Everything else is a percentage of sales.

   Then walk it

   1. Start from the drivers, not the statements. For a subscription business that is subscribers, ARPU and net retention. For a retailer it is store count, sales per square foot and gross margin. Those get real builds with monthly or segment granularity.
   2. Everything else gets a ratio: SG&A as a percentage of sales, D&A off a simple schedule, working capital on days, capex as a percentage of sales, tax at the guided rate. Resist the urge to build a full three-statement cascade for a name you are screening.
   3. Standardise the template across the coverage universe. Same rows, same order, same colour convention for inputs, same output block. Then updating twenty models after earnings is a mechanical exercise, and you can compare names line by line without re-reading each file.
   4. Tier the coverage explicitly. Five or six core names get deep models with segment detail and a channel-check overlay; twenty to thirty monitored names get a driver model with consensus alongside; the rest get a screen. Coverage breadth comes from the tiering, not from making every model thinner.
   5. Build the comparison in rather than bolting it on. Every model should show consensus next to my numbers and the implied valuation at a range of multiples, because the output I actually need is the gap versus the street, not a standalone forecast.
   6. The limitation to state: a simplified model will miss the thing that was in the footnote, so the trade-off is real. I manage it by re-reading the filings on the core names properly and accepting that on tier three I am running a screen, not a thesis. Pretending a thin model is a deep one is how people get caught.

   Where candidates lose it

   Answering 'keep it simple' with no decision rule. The interviewer wants the criterion you use to choose what gets detail. The materiality test, the driver-versus-ratio split and the tiered coverage model are the substance. And say the cost of simplification honestly, because at a platform you will be asked to cover more names than you can model deeply.

   Expect next

   • How many names can you genuinely cover properly?
   • What would you always model in detail regardless of the sector?
   • How do you update twenty models in an earnings week?

   Reported by candidates at Balyasny Asset Management (Equity Research, New York, 2026). Source: Wall Street Oasis.

3. 078Here are the financial statements of three unnamed companies. Work out what kind of business each one is.Compliance and research processHardcase studyHPS Investment PartnersSpecial Situations · London · 2021

   Say this

   I would read four things in order and narrate as I go: the asset side of the balance sheet, the shape of the cost structure, the working capital cycle, and the capital intensity. Those four together identify a business model within a couple of guesses, and the reasoning is what is being graded, not the final label.

   Then walk it

   1. The balance sheet tells you most of it. Heavy PP&E means manufacturing, utilities, telecom or hotels. Heavy inventory with no PP&E means a retailer or a distributor. Almost no assets but large receivables means services or consulting. Large intangibles and goodwill means an acquisitive or a software business. A balance sheet dominated by financial assets and matched liabilities means a bank, an insurer or a lender.
   2. Then margins and their shape. Gross margin above 70 with heavy sales and marketing is software. Gross margin in single digits on huge revenue is distribution, commodity trading or grocery. High EBITDA margin with heavy depreciation is infrastructure-like, so telecom, towers, pipelines.
   3. Then the working capital cycle, which is the most diagnostic single item. Negative working capital with large payables and fast inventory turns is a supermarket or a restaurant chain. Large deferred revenue is subscription software. Long receivables and inventory days is heavy industry or project work. Receivables that are the business are financial services.
   4. Then capital intensity and leverage. Capex above 15 percent of sales with high depreciation says utility, telecom or semiconductor fab. Very high leverage with stable margins says regulated or contracted cash flows. High leverage with cyclical margins says a leveraged buyout.
   5. Then cross-check with one specific tell per hypothesis. A retailer has operating leases now capitalised as right-of-use assets. An insurer has technical reserves. A hotel has both heavy PP&E and high operating leverage. Say the tell you are looking for before you look for it.
   6. Then commit and quantify your confidence: 'Company A is a grocery retailer, and I am confident because of negative working capital, 25 percent gross margin, 3 percent EBIT margin and inventory turning in under 30 days. If I am wrong, it is a food distributor, and the way to tell them apart is store-level assets versus warehouse assets.' Naming the alternative and the discriminating test is what a credit interviewer is looking for.

   Where candidates lose it

   Guessing early and then defending it. This is a pure reasoning exercise: narrate the evidence in order and let the conclusion fall out. Also, do not neglect the working capital cycle, which is more diagnostic than the income statement. And always name your second-best hypothesis and the test that would separate them.

   Expect next

   • Which of the three would you lend to, and on what terms?
   • Which has the most operating leverage?
   • What single extra disclosure would you ask for?

   Reported by candidates at HPS Investment Partners (Special Situations, London, 2021). Source: Wall Street Oasis.

4. 079Explain the construction of a factor. Why that method, and how would you optimise it?Quant and systematicHardtechnicalACAQR Capital ManagementInvestment Research · New York · 2021

   Say this

   Take value as the example. Define the signal, in this case book to price or a composite of several value measures; clean and winsorise it; standardise it cross-sectionally within industry; then build a long-short portfolio from the ranks, usually top minus bottom quintile, weighted and rebalanced on a defined schedule. Every one of those steps is a choice, and the choices matter as much as the signal.

   Then walk it

   1. Signal definition first, and use a composite rather than a single ratio. Book to price, earnings to price, cash flow to price and sales to enterprise value capture the same idea with different noise, so the average is more robust than any one. That is the main argument for composites over single metrics.
   2. Then the cleaning: point-in-time data with the correct reporting lag so you are not using numbers before they were published, delisted returns included so you are not survivorship biased, winsorise or rank-transform the outliers, and handle negative book values explicitly.
   3. Then neutralisation. Standardise within industry, because a raw value screen just buys banks and sells software. Neutralise size too, or the factor becomes a small-cap bet. The choice of what to neutralise defines what the factor actually measures.
   4. Then portfolio construction: quintile or decile spreads, equal weight versus value weight, rebalance monthly or quarterly. Equal weight shows a stronger factor premium and is much harder to trade. Say that trade-off out loud, because it is where academic factors and investable factors part company.
   5. On optimisation, define the objective honestly: maximise net-of-cost information ratio, not gross return, with constraints on turnover, capacity and exposure to other factors. Then use cross-validation across time and across regions rather than optimising a single sample.
   6. And say the limitation before being asked, because this is the real question inside the question. With enough parameters you can produce any backtest you like. The defences are economic priors before data mining, a small number of specification choices, out-of-sample and out-of-region testing, sensitivity analysis showing the result is not knife-edge, and a documented count of how many specifications you tried. A factor that only works with one lookback and one weighting scheme is a coincidence.

   Where candidates lose it

   Describing the signal and skipping the construction choices. Neutralisation, point-in-time data and the equal-versus-value weighting decision are where the real work is. And on 'how would you optimise it', a candidate who does not immediately raise overfitting has failed the question at a firm built on factor research.

   Expect next

   • How would you know you had overfitted?
   • Why neutralise by industry?
   • How would you test whether your new factor is distinct from momentum?

   Reported by candidates at AQR Capital Management (Investment Research, New York, 2021). Source: Wall Street Oasis.

5. 081Two series can be negatively correlated within each month but positively correlated over a full year. How?Quant and systematicHardtechnicalSCSquarepoint CapitalHedge Fund · Montreal · 2024

   Say this

   Because correlation measured within groups and correlation measured across the pooled data answer different questions. If both series share a common upward trend across months, the between-month variation is positive and can dominate the negative within-month relationship. It is Simpson's paradox in a time series.

   Then walk it

   1. Decompose the covariance into within-group and between-group parts. Total covariance equals the average within-month covariance plus the covariance of the monthly means. Those two terms can have opposite signs, and whichever has more variance wins the pooled number.
   2. Concrete picture: every month, A and B move in opposite directions day to day, so within-month correlation is negative. But each month both drift higher, so the monthly averages rise together. Pool the daily data over a year and the shared drift dominates.
   3. The generic driver is a common slow-moving factor. Both series load positively on something persistent, such as inflation, liquidity or a market trend, while their high-frequency innovations offset. Long-horizon correlation is dominated by the common factor and short-horizon correlation by the idiosyncratic part.
   4. There is also a pure measurement version of this: correlation of returns is horizon dependent when returns are autocorrelated. Compute correlation on daily returns and on annual returns for the same pair and you generally get different numbers, and neither is wrong.
   5. Why it matters practically, which is what the interviewer is really testing: hedge ratios and diversification estimated at one horizon do not hold at another. A pair that looks hedged on daily data can be a directional bet over a year, which is exactly how a relative value book acquires an unintended factor exposure.
   6. So the answer to 'which correlation is right' is neither. You choose the horizon that matches your holding period and your rebalancing frequency, and you look at both to know which part of the relationship you are actually trading.

   Where candidates lose it

   Treating it as a paradox to be resolved rather than a decomposition to be stated. Write down the within-plus-between covariance split and the answer is immediate. And do not stop at the maths: the reason they ask is the practical consequence for hedge ratios at different horizons.

   Expect next

   • Which correlation would you use to set a hedge ratio?
   • How does return autocorrelation affect measured correlation?
   • Give me another example of Simpson's paradox in markets.

   Reported by candidates at Squarepoint Capital (Hedge Fund, Montreal, 2024). Source: Wall Street Oasis.

6. 082Talk me through your research process for a systematic signal. How do you avoid fooling yourself?Quant and systematicHardtechnicalBalyasny Asset ManagementQuantitative Trading · London · 2025

   Say this

   Start with an economic reason the signal should work, then test it in a way that can fail. Hypothesis first, then data preparation, then a simple specification, then out-of-sample and out-of-region validation, then costs, then capacity. The discipline is that the hypothesis comes before the backtest, not after it.

   Then walk it

   1. State the economic mechanism first and write it down before running anything. Who is on the other side, and why are they there? A signal with no story about who is losing money is almost certainly a data artifact.
   2. Then the data work, which is most of the time and all of the risk. Point-in-time data with correct reporting lags, restated figures handled properly, delisted and merged companies included, corporate actions adjusted, and survivorship bias eliminated. Look-ahead bias is the most common silent killer and it always flatters the result.
   3. Then the simplest possible specification. One parameter, no optimisation, sensible defaults. If the effect does not show up in the naive version, it probably is not there. Elaboration after validation, never before.
   4. Then validation that can actually fail: hold out a period you never look at, test in other regions and other asset classes, test across sub-periods, and check that the result is not driven by a handful of stocks or one month. Report the number of specifications you tried, because a t-statistic loses its meaning after the twentieth attempt.
   5. Then costs and capacity, which kill more signals than statistics do. Model spread and impact, compute net-of-cost performance at realistic size, and check whether the signal survives a one-day implementation lag. A signal requiring same-second execution is not a signal for a fundamental-horizon book.
   6. Then the honest self-checks: decide the kill criteria before the test, keep a research log of everything tried including the failures, and have someone else reproduce the pipeline. The uncomfortable truth is that most published anomalies do not replicate out of sample, so my prior on my own new signal should be low.

   Where candidates lose it

   Describing a backtest rather than a research process. The order of operations is the answer: hypothesis, then data hygiene, then a naive test, then validation, then costs. A candidate who does not mention point-in-time data, look-ahead bias and the multiple-testing problem will not get through a quant research interview.

   Expect next

   • How many specifications did you try on your last project?
   • How do you handle restated financials in a backtest?
   • What is your kill criterion for a signal?

   Reported by candidates at Balyasny Asset Management (Quantitative Trading, London, 2025). Source: Wall Street Oasis.

7. 087Why is long-short so much harder to run in India than in the US?India and Category III AIFsHardtechnicalIndian hedge fundsCategory III AIFs

   Say this

   Because the short side barely exists in the cash market. There is no developed stock lending market, so almost all shorting happens through single stock futures, which restricts you to the names in the derivatives segment and imposes roll costs and position limits. The short book is therefore structurally narrower than the long book.

   Then walk it

   1. Start with the mechanics. India has no meaningful securities lending and borrowing depth. The SLB platform exists but volumes are thin, so a practical short is a single stock future, and only a limited set of names, roughly 200 or so, are in the futures segment at any time.
   2. That is the binding constraint. You can be long any of the two thousand tradable names and short only the large and liquid ones, so a classic pair trade in mid caps is frequently not expressible.
   3. Then the cost. Futures roll every month, and the basis moves with borrow demand and positioning, so the cost of a short is not a quoted borrow fee but a roll cost that can widen exactly when you most want the position. There is no term certainty.
   4. Then the position limits and margins. Exchange-level and client-level open interest limits, mark-to-market margin on the futures leg, and periodic regulatory tightening of derivative rules mean the short book has an operational overhead the long book does not.
   5. Then market structure. Retail and derivatives volumes dominate, index options turnover is enormous relative to cash equity, and domestic institutional flows into equity funds are steady and large, so the market has a persistent upward bias that penalises short books.
   6. So what actually works in India looks different: long-biased funds with an index hedge, cash-futures arbitrage, merger and event situations, and long-short expressed within the futures universe. Honest framing: India is a great market for long-biased alpha and a hard market for a market-neutral book, and most successful domestic Category III strategies reflect that rather than fight it.

   Where candidates lose it

   Answering with generic emerging market caveats about liquidity and governance. The specific, correct answer is the absence of a cash stock lending market and the resulting dependence on single stock futures, with all the universe, roll and limit consequences that follow. Naming the futures universe constraint is what shows you know the market rather than the idea of it.

   Expect next

   • How would you hedge a mid-cap long that has no future?
   • What is cash-futures arbitrage and why is it popular in India?
   • What would change if SLB volumes grew?
8. 089What SEBI rules would you need to know before running a long-short book in India?India and Category III AIFsHardtechnicalIndian hedge fundsCategory III AIFs

   Say this

   Four bodies of rules: the AIF Regulations that govern the fund itself, the derivative and position limit framework that governs how you short, the insider trading regulations, and the disclosure thresholds on large positions. Plus the operational rules on valuation, reporting and custody.

   Then walk it

   1. The AIF Regulations of 2012 first: registration, the Category III leverage limit expressed as gross exposure not exceeding twice NAV, the 1 crore investor minimum, the manager's continuing interest, periodic reporting to SEBI, and the requirement to disclose the leverage and risk framework in the placement memorandum.
   2. Then the derivatives framework, because that is how you short. Which names are in the futures and options segment, market-wide and client-level position limits, margining including SPAN and exposure margin, and the periodic tightening of index option rules. Your short capacity is defined by these, not by your conviction.
   3. Then insider trading, which is SEBI's Prohibition of Insider Trading Regulations of 2015. Unpublished price sensitive information is the Indian formulation, and the regime requires a code of conduct, a structured digital database recording who received what information, trading windows and pre-clearance. The structured digital database requirement is a genuinely distinctive Indian feature worth naming.
   4. Then disclosure. Takeover Regulations require disclosure at 5 percent and on changes of 2 percent thereafter, which matters for a concentrated book. Short positions must also be disclosed to the exchanges under the framework that prohibits naked short selling; institutional investors cannot square off intraday.
   5. Then the operational layer: valuation policy and independent valuation of unlisted holdings, custodian requirements, benchmarking of AIF performance, and the compliance test report the manager files.
   6. The honest caveat: this is an actively changing rulebook, with amendments most years on derivative limits, disclosure and AIF structuring. So I would give you the framework and say that the specific thresholds need checking against the current circulars rather than quoting them from memory as though they were fixed.

   Where candidates lose it

   Bluffing specific numbers. Indian regulation changes frequently and an interviewer who works under it will know when a threshold is stale. Give the four buckets confidently, name the distinctive items such as the structured digital database and the no-naked-shorting rule, and flag that the exact limits need verification.

   Expect next

   • What is a structured digital database and who has to keep one?
   • Can an institutional investor square off a short intraday in India?
   • How does the 5 percent disclosure threshold affect a concentrated book?
9. 098Should there be a tax on happiness?Career and fitHardsuperdayBridgewater AssociatesEquity Hedge · Westport · 2025

   Say this

   No, and the reason is measurement and incentives rather than fairness. Taxes need an observable, verifiable base; happiness is self-reported, so any tax on it would be gamed instantly and would punish exactly the behaviour a society wants more of. But the interesting question underneath is whether we should tax consumption that buys status rather than wellbeing, and there I would say yes.

   Then walk it

   1. Take the question seriously and state your reasoning structure before your conclusion. That is the whole test at a firm that asks this: they want to watch you think, not hear an opinion.
   2. Define terms first. A tax needs a base that is observable, measurable and hard to misreport. Happiness fails all three, so the practical objection precedes the philosophical one.
   3. Then the incentive argument. Taxing an outcome discourages producing it. If happiness is partly a product of effort, relationships and choices, taxing it penalises those. Compare with a Pigouvian tax, which we levy on things with negative externalities; happiness has positive ones.
   4. Then the steelman, because refusing to engage with it is the failure mode. There is a real argument that positional consumption imposes an externality: if my spending raises the bar for everyone's sense of adequacy, it makes others worse off, and that is a textbook case for a tax. Progressive consumption taxation is the serious version of this idea.
   5. Then the distributional point. The declining marginal utility of income already underpins progressive taxation, which is arguably a rough approximation of taxing the capacity for happiness that money buys. So a version of this already exists and is defensible.
   6. Then conclude with your view and the condition that would change it. I would not tax happiness; I would tax positional consumption. And if happiness became genuinely measurable and non-gameable, I would revisit the measurement objection but not the incentive one. Then invite the disagreement, because at a firm built on radical transparency, arguing back well matters more than being right first.

   Where candidates lose it

   Treating it as a joke, or giving a confident opinion with no reasoning structure. The firm asking this is explicitly testing how you handle an abstract question in a probing conversation. The other trap is refusing to commit: 'there are arguments on both sides' with no conclusion is the worst answer. Build the argument, take a side, and defend it while genuinely updating if the counterargument is better.

   Expect next

   • What if happiness could be measured perfectly?
   • So what should we tax instead, and why?
   • You have argued for one side. Now argue the other.

   Reported by candidates at Bridgewater Associates (Equity Hedge, Westport, 2025). Source: Wall Street Oasis.

10. 100You have read about our culture. Do you actually think you would align with it?Career and fitHardsuperdayBridgewater AssociatesGeneralist · Westport · 2024

    Say this

    Give a real assessment rather than an endorsement. Name the specific practices, say which ones fit how you already work with evidence, and name the one you would find hardest. A candidate who says they align with everything has not understood a culture built on saying uncomfortable things accurately.

    Then walk it

    1. Be specific about what the culture actually involves rather than quoting the slogan. At a firm built on radical transparency and believability-weighted decision making that means recorded meetings, direct critique in front of others, written assessments of your weaknesses, and disagreement escalated rather than smoothed over.
    2. Then match it to evidence from your own behaviour. Not 'I welcome feedback', which everyone says, but an instance: a time you were told publicly that your analysis was wrong, what you did in the following week, and what changed in your work.
    3. Then the harder direction, which is the better test. A time you told someone senior they were wrong, how you did it, and what happened. Cultures like this need people who will speak up, not just absorb criticism gracefully.
    4. Then name the genuine friction honestly. For most people it is being critiqued publicly without becoming defensive, or having to give hard feedback to someone you like. Pick the one that is true for you and say what you would do about it.
    5. Then say why the trade-off is worth it in your view, in investment terms. Investing is a business where being wrong is normal and expensive, so a process that surfaces errors fast is a competitive advantage rather than a management fashion. That framing shows you understand why the culture exists.
    6. Then leave the conclusion appropriately open: I think I would fit, the part I would have to work at is X, and the honest way to find out is a conversation like this one. Overclaiming certainty about your own fit is itself a failure of the transparency test.

    Where candidates lose it

    Praising the culture. This question is the culture in action, so an unqualified endorsement demonstrates the opposite of what it asks for. Name a specific practice you would find difficult and back your claims with two examples, one of receiving hard feedback and one of giving it. Candidates who cannot produce the second one are not convincing.

    Expect next

    • Tell me about a time you told someone senior they were wrong.
    • Which part of how we work would you struggle with most?
    • What would make you decide this was not the right place for you?

    Reported by candidates at Bridgewater Associates (Generalist, Westport, 2024). Source: Wall Street Oasis.

Firm tags come from public, anonymous candidate reports on Wall Street Oasis: strong signal, not sworn testimony. Firms are named as the places a question was reported, not as partners of Fin Maverick. Answers are written for this page to show how to think out loud; they are not scripts to recite.