Painted Block: Cubes with Red Faces
Expected red faces on random painted cube is a medium quant interview question on Expected Value, reported to have been seen at Hudson River Trading and Jane Street.
MyQuantPartner is not affiliated with, endorsed by, or sponsored by these companies, and all trademarks belong to their respective owners.
This classic quant interview question is about understanding how a geometric transformation affects a random discrete choice. A large three-dimensional object is altered, broken into many smaller components, and you must analyze the random properties of one such component after that transformation. It blends geometry, combinatorics, and probability in a way that feels very natural for trading and risk modeling.
It trains your ability to work with expected value in structured settings, to formalize randomness with indicator variables, and to exploit symmetry and classification of cases without brute force. You practice counting, working with ratios, and reasoning clearly about different types of units that arise from a regular partition.
This matters for quant prep because trading firms love questions where a seemingly messy configuration hides a clean probabilistic structure. In real quant interviews, especially at top firms, you must show you can model a system, identify key classes, and compute expectations reliably and quickly.
What it tests
This problem class is governed by the linearity of expectation and the symmetry of spatial arrangements. When an object is partitioned into smaller units, and a property (like being painted) is distributed over its boundary, the expected value of that property on a randomly chosen unit can be computed by summing the probabilities that each sub-feature (such as a face) has the property, regardless of dependencies between them. The key is to recognize that, although the units are not all equivalent (corners, edges, faces, interiors), the expectation can be decomposed into the sum of probabilities for each feature being on the boundary. This principle holds because expectation is additive over indicator variables, and the probability that a random unit's feature is on the boundary depends only on the ratio of boundary features to total units. Thus, the problem reduces to careful counting and grouping by symmetry classes, not by enumerating all possible cases.
Practise this question with written feedback, or hear it in a spoken mock interview.
Get started free