phone screen followed by a full loop in San Francisco in HQ. It was full day with 5 interivews. I spent the night at a hotel close by and was flown out the next day. The interview process was as expected with some white board coding and design questions as well as a break for a very delicious lunch.

Recruiter reached out to me via LinkedIn. 1 brief conversation and 2 technical phone screens later, I was brought onsite. The candidate gets to dictate how long/short the process needs to take; in my case, I decided to let it span over a period of 1.5 months. Phone screen questions were relatively straightforward: PhoneScreen1: 1. Add 2 large numbers represented as linked lists and return the resulting linked list 2. Check if a binary tree is a valid BST PhoneScreen2: 1. Implement a caching layer for a web application 2. Talk about your favorite unix utility (i spoke about grep, where it shines, and what its limitations are) Onsite Interview: Round1: Implement a scalable service that counts the frequency of words present in tweets. Round2: Talk about an exciting personal cs project. You are given a 2D integer matrix and the "contours" of a space inside that matrix. Count the sum of all cells inside that space within O(1) time using some clever pre-processing. Round3: Join the siblings of a binary tree that does NOT have any parent pointers. Round4: Design a scalable architecture that broadcasts tweets from can account for heavy-hitters (celebrities) and normal users with fewer followers. Implement merge sort for a singly linked list. Round5: I don't remember the question clearly, but it was along the lines of an array with several elements and needing to create a high-value "chain" by connecting an element with other adjacent elements that are at least 1 hop away from each other. For example: consider an array like [12,5,14,15,8,10]. We can form an allowed chain by combining 12, 14, 8 (total weight is 34). But the best chain to be formed here is 12, 15, 10 (total weight is 37).