Kickdrum

There were 3 rounds in total. Round 1: They provided the codebase of their own already implemented application and asked me to add a new functionality to the existing system. The first task was to analyze and understand the codebase, explain the logic, API calls, complete flow, and the changes that would be required. The interviewer expected a detailed explanation of the approach and implementation plan before any coding. After that, I was asked to implement the feature and write the actual code. The interview lasted for around 1 hour, and only the last 15 minutes were given for coding implementation. The use of AI was allowed, but they were monitoring everything through a live full screen share, and there was no clear guidance on how much AI usage was acceptable. I explained the logic, flow, and implementation plan clearly to the interviewer. For coding, I used AI mainly to generate the actual implementation based on the plan I had already prepared. They were looking for someone who could quickly understand an existing codebase and independently add new features and functionality. Their expectations were quite high. Out of 8 people shortlisted for Round 1, only 1 person made it to Round 2. My suggestion would be to focus heavily on understanding existing codebases and practice adding features to them. Also, try to write as much code as possible on your own. Round 2 was a pure LLD and System Design round. Round 3 was an HR and Behavioral round. In the end, they did not select anyone from the entire college. OA Pattern: They asked 20 MCQ questions in 30 minutes and 2 DSA questions with 1 hour 30 minutes to solve them.