Druva Interviews

I applied for the Software Engineer position at Druva through an on-campus drive in September 2021, as a Final Year CSE undergrad targeting a new grad role for 2022. The eligibility criteria required a CGPA above 7 and no active backlogs.

The interview process began with Round 0: Online Assessment on the Mettl Platform, which lasted 90 minutes. It featured MCQs on Data Structures, Computer Networks, and C++, alongside three coding questions (one easy, two medium). The coding questions carried a higher weightage. I managed to solve two coding questions and several MCQs, which led to my shortlisting along with 42 other students.

Round 1: Technical Interview
This round, lasting approximately 50 minutes, was conducted by a Senior Principal Engineer. After introductory exchanges, the interviewer presented two coding problems for me to solve on a Google Doc. I could use any programming language.
1. Count Occurrences of an Element in a Sorted Array: Given a sorted integer array with duplicates and a target integer X, I needed to return the count of X in the array in O(log(n)) time. For example, if arr = [1,2,3,3,3,3,3,3,3,4,4,5,5,6] and X = 3, the output should be 7. I explained my approach using lower bound and upper bound, and due to time constraints, I was asked to implement just the upper bound.
2. Sum of Left Leaves minus Sum of Right Leaves in a Binary Tree: The second problem was to find Sum(left leafs) - Sum(right leafs) for a given binary tree.
With some hints from the interviewer, I was able to solve both problems. We also had a brief discussion about how Google Search works, covering basic concepts like DNS and IP.

Round 2: Hiring Manager Round
This round took place the following day, lasting about 50 minutes, and was led by a Director of Engineering. It was a thorough grilling on my resume, projects, and theoretical subjects. The interviewer started by asking about my three favorite technical subjects and the justifications for my choices. We then delved into specific topics:

• DBMS: Questions included how to retrieve data faster in SQL, what an index is with examples, transaction logs, and how a DBMS handles a scenario where an update to the second table fails within a SQL query involving two table updates. I was also asked if I had designed any database schemas for my projects.
• Resume/Projects: We discussed challenges I faced during my internship and a specific project I had set up from scratch.
• OS: I was asked about threading, a comparison of TCP vs. UDP with real-life examples, which protocols Gmail uses, HTTP and HTTPS differences, what synchronization is with an example, how to achieve synchronization, and the differences between virtual memory and RAM.

Towards the end, I was given an algorithmic question: how to find the 10 most frequently used words in an e-book, similar to the LeetCode problem "Top K Frequent Words". After I provided an initial approach, he asked if any further optimization was possible, and I believe he was expecting a Bucket Sort approach. Finally, a behavioral question about the biggest mistake I’ve made in my life was posed.

Unfortunately, I didn't clear this round, likely because my performance in the theory subjects (OS, Computer Networks, and DBMS) wasn't strong enough.