P40 at Atlassian

How it went

What they would tell you

How to prepare

Their background

Round by round

1. 1

   Karat Screening60 min

   Round 1 was handled by a third-party platform, so it felt a bit different from the usual interviews. It started with some quick system design questions, all back-to-back things like How would you scale a chat service? or What if your database crashes during the busiest traffic? These needed fast answers I didn’t have much time to think. After that, there were 2 DSA questions, both on the easier side, and I handled those pretty smoothly.

2. 2

   DSA60 min

   Task: Implement methods for a content popularity system: increasePopularity(contentId), decreasePopularity(contentId), getMostPopular(). Follow-ups:

   • Increase/decrease by arbitrary values (not just 1).
   • Tie-breaker strategies (e.g., prefer newer content, or content with higher engagement).
   • How to handle concurrent updates? This tested both efficient data structures and reasoning about real-world constraints. I used a priority queue (max-heap) for getMostPopular() and discussed thread safety for updates.
   • Content Popularity System
3. 3

   Code Design (LLD)60 min

   Build an agent rating system: addAgentRating(agentId, rating), getSortedAgentRatingsList(). Follow-ups:

   • Add monthly ratings
   • Get monthly top agents
   • Tie-breakers
   • Concurrency (how to avoid race conditions when adding ratings?) I designed classes for agents and ratings, used a sorted list for monthly leaderboards, and locking strategies (fine-grained locks, optimistic concurrency)
   • Agent Rating System
4. 4

   System Design (HLD)60 min

   In this round, I was tasked with designing a popular K feeds system for Confluence. System requirement

   • Calculate popularity scores for different types of content (e.g., posts, comments).
   • Support querying in windows and updating dashboards in real-time. The interviewer asked me to think different scaling strategies, including load balancing, caching.
5. 5

   Management Round60 min

   Behavioral and situational questions, no coding. Examples:

   • Tell me about a project you owned end-to-end.
   • Describe a time you improved team or system efficiency.
   • When did scope creep occur, and how did you handle it?
   • Have you ever pushed back on a decision? How? During this round, I tried to be honest about challenges, not just successes.
6. 6

   Value Round60 min

   Main Focus: Atlassian’s cultural values teamwork, mentorship, integrity etc. Questions: When have you mentored someone? Describe a conflict with a teammate and how you resolved it. Have you ever shipped a bug to customers? What did you learn? Tell me about a time a project was deprioritized—how did you react? I answered each question with real stories, some of which I felt showed growth and reflection.

What came up