PubMatic Interview Questions
& Process

Applied through a LinkedIn referral. I have 1.3 years of experience. The interview process had 3 rounds. Round 1 involved 2 DSA questions: rotating an array at index k and finding a loop in a linked list. There were also questions on core Java and Spring Boot, covering Spring MVC architecture, scaling a project to a million users, Java Streams, OOPS, database indexing, and resume-based questions. Round 2 had one DSA question about finding the winner in an NxN tic-tac-toe board. Java and Spring questions included multithreading (avoiding race conditions, implementations), bean scope, Singleton design pattern, indexing and B-trees, DFS implementation (pseudo-code), anomaly detection, and the diamond problem in Java. Round 3 was focused on Low-Level Design (LLD) for a snake and ladder game, scaling a system to a million users, project Q&A, and a general discussion about GenAI.

The interview process here was awful- be warned! After having a virtual interview and in person round with the CTV team (who were lovely for what it is worth) I had to hound HR for an update on the status of the role. After having to continuously follow up and hearing nothing, I was told by HR I did not get the job- disapointing ,but it happens. The next day HR reaches out to tell me they made a mistake and that I was still being considered for the role and that I would hear back by the end of the week. After that interaction, I was ghosted entirely, had to reach out again weeks later, only to be told again the role was filled by someone else. This is the most unprofessional HR team I have ever been plagued to have to work with and one of the odder interactions I've had in the job search.

So there were 4 rounds. Round 1 was about coding, automation, and SQL. Round 2 was again coding, SQL, and Linux commands. Round 3 was the same topics as before, it was a techno-managerial round. My code worked and I answered everything right. Round 4 was an in-person meeting at their office, they called it a 'general discussion.' After that, HR told me the panel gave negative feedback, but they weren't clear why. It felt like a waste of my time with all the technical rounds, and then they just dismissed me without a real reason.

This was a long interview process with 4 technical rounds. It felt a bit biased, especially towards the Senior Manager level. It started with Round L1, where the interviewer had about 7 years of experience. They asked me some good DSA and SQL questions. There were 4 questions in total, and I had to code the solutions in an IDE of my choice, with the interviewer testing them against 3-4 test cases. After that, some SQL queries were asked, I remember 3 correctly. Then, basic Hadoop and Spark concepts came up. This round went well, and I answered most questions. Round L2 was with a Principal Software Engineer with 13-14 years of experience, focusing purely on data engineering topics like Spark, Kafka, and optimizations. I answered everything, and the interviewer seemed satisfied. Round L3 was with a senior manager from the US and lasted an hour, focusing on my past projects. Towards the end, they gave me two take-home assignments with good questions. After submitting, I had another session with the same manager. They asked me to run the code on an online IDE and then asked questions about it. It felt like a formality; they didn't seem to understand the code or intend to. They claimed they expected the most accurate solution, but the problem statement asked for the most efficient one. I realized then they weren't moving forward, which was confirmed by the feedback. I suspect they might have already hired someone at a lower cost or the manager had other personal motives.