Land the quant seat in trading or research
Quant is narrow and learnable: clear probability, fast expected-value reasoning, and the discipline to take a structured bet instead of hedging. We build the prep and the calendar that match what these desks actually screen for.
Two roles, one probability core, different emphasis
Trading and research interviews probe the same five abilities. The emphasis is where they split, and prepping the wrong one wastes the runway. Unsure? Start with the shared probability core, it serves either seat.
Think mathematically under pressure
Understand probability deeply, not just formulas
Reason about expected value fast, often in your head
Manage risk logically
Think in edge and robustness, not just being right
Quant Trader
- Emphasis
- Intuition, speed, mental math, probability puzzles.
- Interview style
- Rapid-fire EV games, brainteasers, market intuition.
- The bar
- The mental-math bar is very high. Answers land in seconds, out loud.
Drill first: daily arithmetic out loud, then timed EV games until committing to a number feels normal.
Quant Researcher
- Emphasis
- Modeling, statistical rigor, signal testing.
- Interview style
- Derivations, time-series reasoning, design a signal.
- The bar
- Reasoning depth matters more than raw speed.
Drill first: time-series properties (stationarity, mean reversion) and how you would test for each, plus the backtest traps.
The quant recruiting clock starts before you think it does
Quant has the most brutal calendar of any track. The single most common way a strong candidate loses is missing the window before they ever solve a problem. Here is how the year runs.
- Sophomore fall
Your resume already has to be quant-ready
Competition results, relevant coursework, and any modeling or trading projects belong on the page now, not junior spring. The firms that recruit earliest are reading it first.
- October to November
Early-ID deadlines close
Many top quant firms run early-identification programs that recruit before junior year, with application deadlines as early as October and November. This is the window that quietly ends most strong candidates' chances.
- That same fall
You apply for the following summer
Quant runs a year ahead. For early-ID programs you may be applying as a sophomore for a seat you take as a junior. By the time most students think 'I should start prepping,' the best deadlines have passed.
- If a deadline is near
Submit before you feel finished
A submitted application beats a perfect one that missed the window. Missing the window is not a skill problem, and no amount of probability practice fixes it.
The shared core you must know cold
If any of these makes you pause, drill it until it is automatic. In a rapid-fire round you have seconds, and hesitation reads as weakness. Open each cluster, then mark the ones that already feel automatic.
Where do you stand? Mark each cluster you know cold.
0 of 4 cold
Expected value and variance
EV is the long-run average: sum of outcome times probability. Two bets with the same EV can carry wildly different risk.
The complement rule for 'at least one'
Count the opposite: P(at least one) = 1 minus P(none). At least one head in three flips: 1 minus (1/2)^3 = 7/8.
Conditional probability and Bayes
When the base rate is low, even an accurate test gives a low posterior. Say that insight out loud.
Independence, AND vs OR
Independent AND multiplies. OR adds, then subtracts the overlap.
Normal distribution and the CLT
Averages of many independent variables tend toward normal. It is why so much of finance leans on normality.
Markov processes
The future depends only on the present, not the path that got there.
Martingales
The classic probe: is a price a martingale? The best estimate of tomorrow's price is today's.
Mean reversion vs trend
Identify the property, then say how you would test for it.
Two-digit multiplication, out loud
A trader who fumbles 17 times 23 loses credibility regardless of the math behind it.
Fractions and percentages
Convert fast in both directions, no pencil.
Powers of two
Know them cold; they anchor quick halving, doubling, and approximation.
Fast approximation
Approximate, state the direction of your error, keep moving. No freezing.
Calls, puts, implied volatility
Right to buy, right to sell. Option prices rise with volatility: more chance of a large favorable move.
Delta and gamma
Delta is the option price's sensitivity to the underlying; gamma is delta's rate of change.
Sharpe, drawdown, VaR
Return per unit of risk, largest peak-to-trough loss, worst expected loss at a confidence level.
Return is meaningless without risk
Say it before the interviewer has to. Pair every return with volatility or drawdown.
Assume, then bet: run it yourself
The cardinal sins on weak candidates are always the same: no depth, no structure, jumping between numbers with no formula. The fix is one repeatable habit. State the expression, then fill it in. Try it on the staple EV game below.
- 1
Clarify and assume
State your assumptions out loud. 'I will assume the coin is fair and the draws are independent.'
- 2
Name the quantities
Say which expected value and which risk you are about to compute before you touch a number.
- 3
Decompose
Break the problem into clearly-defined parts and define your variables. Structure first, arithmetic second.
- 4
Compute and commit
Reason to a number, then take the bet. Direction beats neutrality. 'It could be 30 to 70 percent' is a non-answer.
- 5
Stress it
Name the risk and what would change your answer. EV tells you there is edge; variance tells you whether it survives bad luck.
A fair coin. Heads, you win 3 dollars. Tails, you lose 1. It costs nothing to play. Do you play?
- Assume
- Fair coin, independent flips.
- State the expression, then fill it in
- EV = 0.5(3) + 0.5(-1) = +1 per play
- Commit
- Play, every time. The edge is positive.
- Stress it
- Variance is still real: you can lose 1 on a bad flip and several in a row. Over many plays the edge dominates.
In coaching we run dozens of these live, narrated out loud, until the structure is reflexive under pressure.
Coached by people who landed quant roles
You are repped by someone who has already cleared the rounds you are preparing for, not a generalist reading a script. They know what these desks screen for because they sat on the other side of it.
Mentor slot
Placeholder. Real coach credibility (where they landed, school) goes here once supplied. No names tied to the track.
Mentor slot
Placeholder. Real coach credibility (where they landed, school) goes here once supplied. No names tied to the track.
Mentor slot
Placeholder. Real coach credibility (where they landed, school) goes here once supplied. No names tied to the track.
What quant students say
Real, attributed quotes from students who landed trading and research seats will live here. We do not run anonymous or invented testimonials.
Testimonial slot. A specific, attributed quote (name, role, and the outcome it produced) goes here once collected.
Testimonial slot. A specific, attributed quote (name, role, and the outcome it produced) goes here once collected.
What loses the round, and what we fix
Most rejections are not a math gap. They are a habit gap the interviewer reads in seconds. We name the failure mode, then drill the replacement until it holds under pressure.
No structure, jumping between numbers
State the EV or probability expression first, then fill it in.
Hedging into neutrality, will not commit
Reason to an estimate and commit, with direction, then adjust.
Bragging about returns, ignoring risk
Pair every expected value with variance, drawdown, or VaR.
Trusting a too-good backtest
Assume an error first: lookahead bias, overfitting, ignored costs.
Ignoring transaction costs
Model costs and slippage before you call a paper edge real.
Freezing on mental math
Drill arithmetic daily until it is automatic, out loud.
See the structure, state the EV, commit to a number
Quant internships are demanding but attainable for students who pair strong fundamentals with a committed bet and honest risk. Start with a free consult on your background, target firms, and the deadlines that matter.