1. Initial screening interview (couple months back)
   • sample from multinomial array n numbers, e.g [0.5, 0.25, 0.25]
     • Came up with solution that uses binary search, didn't manage to code in time
2. On-site virual interviews
   • Coding interview
     • Given 2 nodes in BST, find deepest common ancestor
       • Came up with working single traversal solution, although the problem was unclear (they had "lowest common ancestor" in the problem description, so remember to ask clarifying questions!)
     • Design add, delete, deleterandom() with O(1) time complexity: designed suboptimal solution which used doubly-linked list, didn't get even hint that I'm going to wrong direction. I've checked later on LeetCode and I've actually done exactly that problem before :((
   • Big Data coding, given sorted array e.g. [1,2,3,4,5], return sorted array after applying following function:
     f(x) = ax^2 + bx + c, a > 0 (so it's a parabola opening upward)
     Designed optimal solution, though with hints and also didn't manage to completely code it in 1 hr
   • Machine Learning interview:
     • 2 itnerviewers asked some NLP based questions, how would I design etc. questions were too vague and didnd't ask anything concretely, after that went to some basic questions about Precision, Recall. I think it could've went smoother.
   • Interview with manager:
     • Behavioral questions etc.

Result: Reject