AI Fin Hub

Generator

Quant Interview Question Generator

Seed 7 draws 8 medium questions, each with a worked solution, from a bank of 45.

On the documented defaults, computed by this tool's engine · the same answer as JSON

Curated bank of probability, stats, derivatives, microstructure, and regression questions across easy/medium/hard. Reproducible by seed.

01Generator

Runs in your browser. Nothing you enter is uploaded, and no account or API key is needed.

Education, not investment advice. Past performance does not predict future results. How we check our numbers.

Configuration

Topics

Difficulty

Number of questions

Fewer appear when the filters match fewer than 8.

Seed

Same seed and filters, same set, so you can share or revisit it.

45 questions in a fixed bank. Nothing is generated live; the seed only picks and orders them.

Questions selected

8

2 probability, 3 statistics, 1 derivatives, 1 microstructure, 1 regression; difficulty: medium.

Question set

  1. Q1, derivatives, medium

    Why is the Black-Scholes price of an at-the-money call approximately 0.4 · S · σ · √T?

  2. Q2, statistics, medium

    You estimate Sharpe = 1.0 over 252 days. Construct an approximate 95% CI on the annualized Sharpe.

  3. Q3, microstructure, medium

    What is adverse selection in an order book?

  4. Q4, probability, medium

    100 people board a plane in order; the first sits randomly, others in their assigned seat if free, else random. P(last person gets their assigned seat)?

  5. Q5, probability, medium

    X and Y are i.i.d. uniform on [0,1]. What is P(|X − Y| < 0.1)?

  6. Q6, statistics, medium

    Why does multiple-testing inflate the false-discovery rate, and what's one classic correction?

  7. Q7, statistics, medium

    What is the bias-variance decomposition of MSE?

  8. Q8, regression, medium

    Why are robust (HAC / Newey-West) standard errors used in financial regressions?

02Receipt

What the answer above ran on, how it is computed, where the data comes from, and a link that reopens it.

Inputs

  • difficultymedium
  • n8
  • seed7

The documented defaults. Change any input above and the tool recomputes.

Method

How the hub's tools are built and checked

Engine /engines/quant-interview-question-generator.js, sha256 416e97780f49. The page, the API and MCP run the same file.

Source

Your inputs

Page updated .

Permalink

aifinhub.io/quant-interview-question-generator/

The link carries the inputs you set, so it reopens this result.

03For agents

Any agent can run this calculator by URL or MCP and get the same answer as JSON: the inputs that used defaults, the result and a line to cite. The call below follows the inputs you set above. In code, import the same engine; it runs where you import it, with no request or key.

Requestthe documented defaults
GET https://aifinhub.io/api/run/quant-interview-question-generator
Response200 application/json · 4.4 KB
{
  "calculator": "quant-interview-question-generator",
  "title": "Quant Interview Question Generator",
  "site": "AI Fin Hub",
  "input": {},
  "defaults_used": {
    "difficulty": "medium",
    "n": 8,
    "seed": 7
  },
  "result": {
    "questions": [
      {
        "id": "d-m-001",
        "topic": "derivatives",
        "difficulty": "medium",
        "prompt": "Why is the Black-Scholes price of an at-the-money call approximately 0.4 · S · σ · √T?",
        "solution": "With K = S and r = q = 0: C = S·[Φ(d1) − Φ(d2)], d1 = σ√T/2, d2 = −σ√T/2. For small σ√T, Φ(d1) − Φ(d2) ≈ φ(0)·σ√T = σ√T/√(2π), so C ≈ 0.3989·S·σ√T ≈ 0.4·S·σ√T."
      },
      {
        "id": "s-m-001",
        "topic": "stats",
        "difficulty": "medium",
        "prompt": "You estimate Sharpe = 1.0 over 252 days. Construct an approximate 95% CI on the annualized Sharpe.",
        "solution": "Lo (2002), i.i.d. returns: SE(SR) ≈ √((1 + SR²/2) / n) with SR and n in the SAME period. Daily: SR_d = 1/√252 ≈ 0.063, n = 252 → SE_d ≈ √(1.002/252) ≈ 0.063. Annualize ×√252 → SE ≈ 1.0, so the 95% CI is roughly 1 ± 2, i.e. about −1 to 3. Equivalently SE(SR_ann) ≈ √((1 + SR_ann²/(2·252)) / years). One year of data cannot distinguish a Sharpe-1 strategy from zero; plugging the annual SR into the daily n (√(1.5/252) ≈ 0.08) is the classic mistake."
      },
      {
        "id": "m-m-001",
        "topic": "microstructure",
        "difficulty": "medium",
        "prompt": "What is adverse selection in an order book?",
        "solution": "Resting limit orders get hit when price moves against them — informed traders pick off stale quotes. Market makers price this into the spread."
      },
      {
        "id": "p-m-002",
        "topic": "probability",
        "difficulty": "medium",
        "prompt": "100 people board a plane in order; the first sits randomly, others in their assigned seat if free, else random. P(last person gets their assigned seat)?",
        "solution": "Classic — 1/2. Symmetry: only seat 1 vs seat 100 ever matter, equally likely to be taken first."
      },
      {
        "id": "p-m-003",
        "topic": "probability",
        "difficulty": "medium",
        "prompt": "X and Y are i.i.d. uniform on [0,1]. What is P(|X − Y| < 0.1)?",
        "solution": "Geometric: area of |X−Y|<0.1 strip in unit square = 1 − (0.9)^2 = 0.19."
      },
      {
        "id": "s-m-002",
        "topic": "stats",
        "difficulty": "medium",
        "prompt": "Why does multiple-testing inflate the false-discovery rate, and what's one classic correction?",
        "solution": "Each test has α false-positive rate; K independent tests → expected K·α false positives. Bonferroni: divide α by K. BH (Benjamini-Hochberg): controls expected FDR at q."
      },
      {
        "id": "s-m-003",
        "topic": "stats",
        "difficulty": "medium",
        "prompt": "What is the bias-variance decomposition of MSE?",
        "solution": "MSE(θ̂) = Bias(θ̂)² + Var(θ̂). Total error trades off systematic miss against estimator volatility."
      },
      {
        "id": "r-m-002",
        "topic": "regression",
        "difficulty": "medium",
        "prompt": "Why are robust (HAC / Newey-West) standard errors used in financial regressions?",
        "solution": "Returns and residuals are heteroskedastic and serially correlated. Vanilla OLS SEs assume i.i.d. residuals → understate uncertainty. Newey-West corrects for autocorrelation up to chosen lag."
      }
    ],
    "totalInBank": 45
  },
  "provenance": {
    "run": "https://aifinhub.io/api/run/quant-interview-question-generator",
    "page": "https://aifinhub.io/quant-interview-question-generator/",
    "methodology": "https://aifinhub.io/quant-interview-question-generator/",
    "contract": "https://aifinhub.io/contracts/quant-interview-question-generator.json",
    "engine": "https://aifinhub.io/engines/quant-interview-question-generator.js",
    "engine_sha256": "416e97780f4952311ce8eedc4ba7406df705c718d34853d1f0c0566f4ddae27d",
    "build": "1a772910d31ef1740892ed0e3a065ed71e8c7406"
  },
  "disclaimer": "Planning estimates only — not financial, tax, or investment advice.",
  "license": "https://creativecommons.org/licenses/by/4.0/",
  "cite": "AI Fin Hub, Quant Interview Question Generator (engine 416e97780f49), https://aifinhub.io/api/run/quant-interview-question-generator"
}
The response for the defaults, as the run URL returns it. Provenance carries the engine hash and build.

How to use it

  1. Pick topics (probability, statistics, derivatives, microstructure, regression) and a difficulty: easy, medium, hard or any.
  2. Set how many questions you want and a seed. The same seed and filters always give the same set.
  3. Work each question without notes, then tick Show solutions to check your answer.
  4. Change the seed for a fresh draw from the bank, or share the seed to give someone the same set.
  5. Use the hard questions to practise explaining a derivation out loud, which is what interviews test.

Questions people ask

What categories does the generator cover?

Five topics: probability, statistics, derivatives, market microstructure and regression, each with easy, medium and hard questions. The bank is fixed and small; the page shows how many questions it holds.

Are these real interview questions?

They are standard problems of the kind used in quant interviews (classic probability puzzles, Black-Scholes and regression derivations, microstructure concepts), written for this bank. They are not copies of any firm's current interview questions.

How do I use this to prep?

Filter to one topic and difficulty, solve each question without aids, then reveal the solution. Move up a difficulty once you can explain every solution in your own words, and change the seed for a new draw.

What's the difficulty calibration?

Easy questions test one definition or calculation, medium ones combine two ideas, and hard ones need a derivation or an argument about assumptions. The labels are editorial judgement, not calibrated against interview data.

Does it grade my answers?

No. Each question has one worked solution to compare against; grading is self-assessment.

  • Playgrounds Calibration Dojo

    Train your probabilistic intuition. Answer binary forecasting questions at any confidence level; track Brier score and reliability curve over time. All state stored locally.

  • Generators Trading System Blueprinter

    Pick your data source, LLM, broker, storage, risk engine, and logger. Get a Mermaid architecture diagram and a copyable starter file tree — the full stack before you write code.

  • Calculators Kelly Criterion Calculator

    Size positions with full, half, quarter, or eighth Kelly and stress-test the choice with a drawdown Monte Carlo simulator. Client-side. Private by default.