Market Making Interview Prep
A guide to market making interview games, fair value, spread setting, inventory risk, adverse selection, and calibration.
Quant trading candidates preparing for market making interviews.
What market making interviews test
Market making interviews test five things at once: fair value estimation, honest uncertainty, spread discipline, inventory management, and quote updates. The interviewer is mostly watching whether you learn from trade flow instead of defending your first number. That is why a candidate who quotes wide, gets picked off once, widens further, and explains why will usually score above a candidate who quotes tight and happens to make money.
Concrete example
You are asked to make a market on the number of coins in a jar. You estimate 800, but you would genuinely not be shocked by 600 or by 1000, so you quote 650 at 950 - wide, and defensible. Quoting 790 at 810 against that same uncertainty is a standing offer to anyone better informed. Now the interviewer lifts your offer at 950. The instinctive read is that you sold well; the correct read is that somebody was willing to pay 950 for a thing you valued at 800, which is evidence your estimate was low. So the midpoint moves up to roughly 850, and you are now short one unit and want to buy it back, which means skewing the whole market upward to make a fill on your bid more likely: 720 at 1020 rather than a symmetric 700 at 1000. Every one of those moves needs a sentence of reasoning attached to it.
How to practice
Use a fixed loop: estimate fair value, state the uncertainty, choose a width that matches it, quote, observe the trade, update, and account for inventory. Log all six fields every time. The discipline that pays is explaining each move - the market widened because the information got worse, narrowed because it got better, skewed because inventory went one-sided, or shifted because the last trade was itself information. If you cannot name which of those four caused a quote change, you guessed.
Common mistakes
Candidates quote too tightly before they understand the distribution, ignore inventory entirely, or move quotes with no stated reason. The deepest mistake is reviewing games on profit and loss, because a lucky win hides bad quotes and a loss can follow from a correct decision under uncertainty. Profit is an output, not a scorecard. Review instead whether the width matched the uncertainty, whether the update moved in the right direction, and whether the inventory was acknowledged at all.
Practice the pattern
Use the LeetQuidity curriculum and calibration to turn this topic into a focused practice plan.