Quant interview preparation: what firms really test

They are scoring the reasoning, not the answer

The single most common mistake is silence. Interviewers are trying to see how you decompose a problem, what you assume, and whether you notice when your own answer stops making sense. A wrong answer explained cleanly outperforms a right answer produced from nowhere.

The question families

• Probability. Conditional probability, expectation, and games where the trick is setting the problem up correctly.
• Statistics. Estimation, bias, variance, and knowing when a result is noise.
• Mental maths. Fast arithmetic and estimation, especially for trading seats.
• Programming. C++ and Python, data structures, complexity, and — for engineering seats — memory and concurrency.
• Brainteasers and market questions. Fewer than reputation suggests, but not zero.
• Your own work. Expect to be pushed hard on one project you claim to know well.

A preparation plan

Work in the order of the seat you want. Trading: arithmetic and probability daily, then market intuition. Research: statistics depth, then your project narrative. Development: C++ fundamentals, then system design.

Do timed practice out loud. The gap between "I know this" and "I can say this at speed while being interrupted" is where most candidates lose the offer.

Practise against real structures

Our interview lab lets you filter original practice questions by topic and by the kind of firm, so you can rehearse the specific mix a given process is likely to use.