The tech screen involved a rapid-fire session: 3 ML-related questions followed by 2 LeetCode medium problems.

540. Single Element in a Sorted Array

We have a near-duplicate pipeline at Pinterest, which produces a mapping from every image to a list of up to k near-duplicate images, such as:

json near_dups = { "A": ["B", "I", "K"], "B": ["A", "D"], "C": ["E"], "D": [], "E": [], "F": [], "G": ["K"], "I": [], "K": [] }

Given a mapping like this, form near-duplicate clusters: collections of almost identical images. In the example above, we would form the following groups: (A, B, D, I, G, K), (C, E), and (F).