Real, anonymized candidate journeys from recent (2025) Paytm SDE and SSE loops — round by round, including what was asked, how the candidate approached it, and the outcome.
One complete journey, start to finish
Round 1 — DSA & Core Concepts
A recruiter called me in early September and the first round was set up for a Saturday morning over video, about an hour. The interviewer started by asking about my current work, and since I had spent two years on a payments team, we spent ten minutes on that — how our transaction flow worked and what broke most often. Then he shared a coding link with the Unique Paths problem: count the ways to reach the bottom-right of a grid moving only right or down. I explained the recursion first, said it was exponential, then wrote the bottom-up DP with a 2D table and dry-ran it on a 3x3 grid. He asked me to justify the time and space complexity, and then whether I could cut the space, so I showed the single-row version. After that he moved to Spring Boot — how dependency injection works and what a bean scope is — and then Kafka: consumer groups, partitions, and what happens when a consumer dies. It ended on time and I heard back two days later.
Round 2 — Java Deep-Dive & Problem-Solving
The second round was three days later, also about an hour, and it was mostly Java. The interviewer went in order: first garbage collection, how generations work and when a full GC happens. Then interfaces versus abstract classes and default methods. Then threading: how ThreadPoolExecutor's core size, max size and queue interact, and what happens when the queue fills up. She asked me to write a thread-safe singleton, and I wrote the double-checked locking version with a volatile field and explained why volatile was needed. Then how to make a class immutable, and I listed the rules — final fields, no setters, defensive copies. After that it moved to databases: optimistic versus pessimistic locking and when I would pick each. Then caching strategies, cache-aside versus write-through, and how a JWT is structured and verified. Near the end she gave me the H-Index coding problem. I sorted the citations and walked from the largest down until the count condition broke, then mentioned the counting-sort O(n) version. We finished right at the hour.
Round 3 — Managerial
The managerial round came about a week after the second round, with the hiring manager, around 40 minutes. He asked me to walk through my experience in my own words, and we talked about the projects I had led and one production issue I had handled end to end. Then he described the team's tech stack — Java, Spring Boot, Kafka, and the databases they used — and asked how comfortable I was with each. I was honest about where I was strong and where I would need to ramp up. After that it was logistics: my notice period, whether I could join within two months, and my location preference. He asked if I had questions and I asked about the team size and how work was divided. The next day the recruiter told me I had cleared all three rounds. Then HR called for the compensation discussion, and the number they came back with was below what I had asked for. We went back and forth once, it did not move enough, and I decided to decline the offer.
Key takeaway: The candidate cleared all three rounds and was selected, but reportedly declined after the compensation discussion with HR didn't work out — be ready to justify DP solutions on complexity, and expect a genuinely deep Java/concurrency/caching grill at this level.
More real interview journeys
The first round of each journey below is shown in full; the rest of each candidate's journey — every remaining round, verbatim — is part of the Placement-Ready PYQ Kit.
Online Assessment (HackerRank, ~63 min)
A recruiter messaged me on LinkedIn in the first week of August and sent the HackerRank link the same evening. The test had a fixed window of about 63 minutes, so I took it that night after work. There were two coding problems and six multiple-choice questions on data structures. I did the MCQs first because they were quick — things like time complexity of heap operations and which structure fits a given use case. Then I opened the first coding problem, a string question. I wrote a brute-force version, checked it against the sample, and then improved it before moving on. The second problem was an array problem with a selection rule; I got the main cases passing and one edge case failed, which I fixed with about five minutes left. I submitted with a minute to spare. Two days later the recruiter wrote back saying I had cleared the assessment and asked for my availability for the technical rounds the following week.
🔒 2 more rounds — continue this journey in the full kit
DSA Online Assessment
I applied through a referral from a college senior and got the online assessment link about a week later. It was two problems. The first one was the sliding-window question: for every window of size k, print the first negative number, or 0 if there is none. I first thought of scanning each window from scratch, then realised I only needed to keep the indices of negative numbers in a queue and drop the ones that fell out of the window. I coded that with a deque, ran it on the sample [12, -1, -7, 8, -15, 30] with k = 3, and got [-1, -1, -7, -15], which matched. The second problem was to implement an LRU cache with get and put in O(1). I wrote a hashmap plus a doubly linked list with dummy head and tail nodes, tested put, get, and an eviction, and it passed all the hidden cases. I finished about ten minutes early. The recruiter confirmed the next day that I had cleared it and set up the design rounds.
🔒 2 more rounds — continue this journey in the full kit
Online Assessment (HackerRank)
I applied on the careers page for a senior frontend-leaning role and got the HackerRank invite about ten days later. The assessment had three parts. I started with the six multiple-choice questions, which were basic data-structure and JavaScript questions, and finished them in a few minutes. Then I did the DSA problems — array and string questions that I solved with straightforward approaches and tested on the samples. The last part was the React machine-coding task: build a small component with a list, a filter input, and some state handling, inside the browser editor. I set up the component structure first, got the data rendering, then added the filter and made sure it did not re-render more than needed. I checked it in the preview pane a few times as I went. I submitted with a few minutes left. About four days later the recruiter emailed to say I had cleared the assessment and scheduled the first technical interview for the following week.
🔒 2 more rounds — continue this journey in the full kit