Google

I applied for the Google SWE Intern role through the off-campus process. After the initial resume screening, I was shortlisted for the coding assessment. Online Coding Assessment The coding round was conducted on HackerEarth and had 2 DSA problems to be solved in 1 hour. The first problem involved repeatedly removing consecutive groups of identical characters from a string and finding the minimum number of operations required to make the string empty. The second problem involved a binary string and required counting the number of distinct decimal subsequences that could be generated from it. Both questions required more than straightforward implementation. I was able to solve one problem completely, while I couldn't finish the second within the given time. I still managed to clear the assessment and was invited for the interview rounds. Technical Round 1 The first technical interview was conducted over Google Meet using a shared coding document. The interviewer gave me a DSA problem involving finding the maximum subarray sum where the first and last elements are equal, with the initial version containing only positive numbers. I first explained my approach and then discussed the time and space complexity before coding it. After I completed the solution, the interviewer modified the problem to also allow negative numbers. I had to reconsider the approach and explain how the change affected the solution. The interviewer also asked about my projects. Since one of my projects was deployed using AWS S3, we discussed some basic cloud concepts and how the deployment worked. The round lasted around 45 minutes. Technical Round 2 The second interview started with a deeper discussion about my projects and some basic cloud-service concepts. The interviewer then gave me a considerably harder DSA problem based on Strange Printer II. The problem involved determining whether a given colored matrix could be generated by repeatedly printing solid rectangular regions, with the restriction that each color could only be printed once. I initially approached it recursively and was able to come up with a solution, but the complexity was exponential. The interviewer then asked me to optimize the solution. I initially thought about treating it as a DP problem, but I wasn't able to identify the intended graph/topological-sort approach during the interview. I spent quite some time trying different ideas, but eventually ran out of time without reaching the expected optimized solution. The interview lasted around 45 minutes. A few days later, I received the rejection update. The complete process was: Resume Shortlisting → OA → Technical Round 1 → Technical Round 2 → Not Selected The main takeaway for me was that Google interviews can move quickly from a relatively familiar DSA pattern to a significantly harder follow-up. Knowing an approach isn't always enough—the ability to recognize the underlying graph, DP or greedy structure under pressure is equally important.