Oracle OCI | SDE 1 (IC-2) | Seattle | Jan 2020 [Offer]

I had completed one year of full-time work at Amazon as a CSA when a friend from my university, who was graduating, informed me that Oracle was hiring SDEs in Seattle, WA. He kindly provided me with the LinkedIn contact of a manager there. I reached out to the manager with my resume, and they promptly set up a phone interview.

Phone Interview

My phone interview involved a coding challenge. I was asked to process a text file containing various log messages and identify the K most frequently occurring ones. I initially thought of using a max-heap, but then realized that a min-heap, limited to a size of K, would be a more space-efficient approach. I presented both solutions to the interviewer, highlighting the benefits of the min-heap.

Onsite Interviews

The onsite part consisted of five rounds, a mix of coding, behavioral, and system design questions.

Round 1

This round presented a matrix filled with 0s and Ds, where 0s represent files and Ds represent databases. The goal was to find the shortest distance for each 0 to its nearest D. I first proposed a brute-force solution with O(N^4) complexity, which involved iterating through the matrix and checking neighbors. However, I soon moved to a better solution using Breadth-First Search (BFS), starting from all 0s simultaneously to update distances. I struggled a bit with the BFS implementation but managed to code it out in the final minutes.

Round 2

I was asked to implement a Binary Search Tree Iterator. As this is a standard tree question, I didn't face much difficulty. I initially thought of an inorder traversal to store all elements in an array, which could then be iterated. The optimized solution, which I also discussed, involved a controlled recursion method to act as a true iterator, as explained in the official LeetCode solution.

Round 3 (Lunch with Manager)

After lunch, there was a short coding exercise: calculating the angle between the hour and minute hands of a clock. I had recently discussed this problem with a friend, which allowed me to quickly provide the solution.

Round 4

This round had two parts. The first coding question involved managing a list of unreserved telephone numbers, implementing a `getRandom()` function for retrieving a random number and a `reserve()` function to assign a number, with critical time complexity requirements. Having practiced a similar problem, I was able to solve it efficiently.

The second part was a system design problem: designing a centralized logging system to collect logs from 100 servers and enhance user experience by reducing log search time. We had only 15 minutes for this, and my answers were structured around concepts from a popular system design primer.

Round 5

The final technical round asked me to find the longest repeated or duplicated substring within a given string `s`. This was a hard problem, and I could only come up with a brute-force solution. We touched upon advanced concepts like Trie and the Rabin-Karp algorithm, and the interviewer seemed content with our discussion.

Behavioral Questions

Oracle Cloud Infrastructure (OCI) has its own set of values, quite similar to AWS Leadership Principles. I was expected to share stories demonstrating how I embodied these values. Additionally, the 'bar raiser' interviewer asked detailed questions about projects I completed during my Masters.

By the end of the week, the manager reached out and informed me that I would be receiving an offer.