Atlassian interview guide¶
Builds Jira, Confluence and Trello; interviews reward runnable, tested code in your own IDE, a code design round, and a values interview. Updated October 2026.
| Atlassian at a glance | |
|---|---|
| Category | Big Tech |
| Intern level | Software Engineer Intern (no P level). Summer internship for penultimate-year students; region calendars differ (US May to Sep, India May to Aug, ANZ Nov to Feb). |
| New grad level | P30 (Levels.fyi label 'Junior Engineer'; posted as Graduate Software Engineer / SWE-1). New grads join the 12-month Grad++ program. |
| 0 to 3 years | P30 for roughly 0 to 2 yrs; P40 (Software Engineer II, Levels.fyi label 'Engineer') for roughly 2 to 5 yrs. Hiring committee sets the final level and downlevels are common. |
| Online assessment | HackerRank (India 2026-grad internship drive screenshot); an Atlassian-branded timed IDE with sections was used for the 2027-grad OA (Aug 2026). UK graduate track started with psychometric tests (Sep 2026). : 2 to 4 coding questions, about 90 min (US intern 2025); India campus OLT 2 to 3 questions of roughly Codeforces 1600 difficulty (DP, graphs, bitmasking, greedy, XOR). Sydney: 2 medium plus 1 easy. Data engineering intern OAs mix SQL and LeetCode. |
| Coding rounds | Intern and new grad: 1 to 2 coding interviews (45 to 60 min). P30/P40 experienced loop: 2 coding interviews of 60 min (Data Structures + Code Design), after a Karat or technical screen. |
| Behavioral | Atlassian's five values: Open company, no bullshit; Build with heart and balance; Don't #@!% the customer; Play, as a team; Be the change you seek. Answer in STAR format. |
| Timeline | Official: interviews are scheduled in stages and the recruiter confirms progression after each; a hiring committee decides after the loop; team matching happens after the hiring committee (centralized hiring, not team-first). Reported: US intern about 2 months (Sep to Nov 2025); Sydney intern 3 to 4 months; P40 backend about 2 months from recruiter reply to HC approval, including almost a month of year-end HC wait (Jan 2026; location not stated); P50 loop Dec 6 2025 to Jan 9 2026 with HC verdict Jan 20. Recruiting seasons per region on the early careers page. CHANGE: Atlassian announced layoffs of about 10% of its workforce in March 2026 (Wikipedia); 2026 LeetCode posts describe paused team matches and a re-test via the AI-assisted round for a pre-layoff hire. |
| New grad pay | Levels.fyi (Software Engineer, read 2026-10-04): US P30 average total comp $167,059/yr (base $125,340, stock $32,075/yr, bonus $9,644; 10 data points, recent offers $160k to $168k); US P40 average $248,629. India P30 average about INR 4.49M/yr (about 45 LPA; base about 2.97M); India P40 about INR 6.86M. Australia P30 average about A$142k (base about A$116k). LeetCode posts quoting '75 LPA' (2027 grad) and '85 LPA' (2026 grad) CTC likely count joining bonus and multi-year stock, so compare per-year figures. |
| Official links | Careers, Students, Official interview prep, Values |
Interview process¶
New grad¶
- Application. Apply via the Interns/Graduates filter on the careers site. Rolling review. Official recruiting windows: US and Canada Sep to Mar (grad start July), India Jul to Dec (grad start July), ANZ Feb to Jun (grad start February).
- Assessment (mandatory). Official early careers guide: SWE candidates take an online coding test and must clear it to interview. Reported formats: HackerRank, about 90 min, 2 to 4 DSA questions (India 2026 grad drive had at least 4, incl. 'Count Bitonic Subsequences'); 2027-grad India OA (Aug 2026) asked a weighted 'make array distinct' problem. UK SWE Graduate (Sep 2026) reported psychometric assessments as round one.
- Recruiter / eligibility screen. About 15 min: standard eligibility checks and motivation questions, followed by 2 technical rounds (Graduate Software Engineer review, Aug 2025; location not stated in the review).
- Interview loop (3 to 4 virtual interviews). Official guide: craft interviews (technical) plus leadership and values interviews, STAR format. For SWE grads this is usually 1 to 2 coding interviews of 45 to 60 min (DSA in your own IDE, sometimes a code design problem), a values interview (45 min) and sometimes a manager/leadership interview. A P30 (SWE-1, India, Sep 2026) report shows a two-part DSA problem (longest constant-difference streak, then allow one replacement).
- Hiring committee and team match. A hiring committee reviews all feedback independently and decides (official). Team matching follows; 2025 to 2026 reports describe P40/backend candidates waiting weeks for team match or being put on hold until headcount opened (Sep 2025 backend hire; Jan 2026 P40).
Intern¶
- Application. Penultimate-year Bachelor's or Master's students in CS or related fields; rolling review, apply early. About 300 to 500 interns a year is a third-party figure (unverified).
- Online assessment. HackerRank-style OA. Reports: US intern OA of 90 min with 2 or 3 questions (applied Sep 2025, SF); 2 LC mediums (US intern, interviewed Nov 2024); 3 LeetCode questions (SF intern, 2025 review); Sydney 2026 OA described as 'reasonably balanced in difficulty'; India campus OLT 2 to 3 questions around Codeforces 1600 (DP, graphs, bitmasking), with 2+ solved needed to shortlist.
- Technical interview (45 to 60 min). Live coding with one engineer over Zoom: LeetCode easy-medium with emphasis on approach, clarity and communication (rank-teams-by-votes style hash map voting, LCA, LRU cache, intervals). NEW in 2026: several intern loops (Sydney Apr to May 2026; Guwahati and Kharagpur campus drives Aug 2026) used a 1-hour AI-assisted coding interview in a HackerRank IDE, e.g. fixing a bug in a small backend web app with an AI agent. Frontend interns build a UI component (Sydney, 2025).
- Values / behavioral interview. 30 to 45 min with an engineer or engineering manager, questions mapped to each Atlassian value; India campus loops call this the HR or 'LNV' round. Some loops add a recruiter call before the final round.
- Offer. US intern timeline example: applied Sep 9, OA Sep 20, round 1 Oct 18, round 2 Nov 3, offer call Nov 9 (2025 cycle). Sydney candidates report 3 to 4 months end to end.
With 1 to 3 years of experience¶
For 1 to 3 yrs (P30/P40) the loop is staged. Screen: Karat third-party interview (60 min: about 20 min of 5 rapid-fire system design scenarios plus 2 easy-medium DSA problems), reported for P30 and P40 in 2025; an Oct 2026 candidate was told the technical screen is 20 min coding plus 15 min HLD. Stage 1: 2 coding interviews of 60 min each, Data Structures (own IDE, tests, debugger allowed) and Code Design. Official prep PDFs dated Jul and Aug 2026 say Code Design is now AI-enabled: a HackerRank IDE with AI coding assistants and an existing code repository you extend, fix or refactor, in a language you may not know. Stage 2: 60-min system design (ladder of follow-ups). Stage 3: 60-min management interview (own outcomes, expertise-driven improvements, initiative, team contribution) and 45-min values interview. Each stage is gated by recruiter feedback. Hiring committee then levels you; candidates report being downleveled P50 to P40 for skipping clarifying questions in design.
Online assessment¶
- Platform: HackerRank (India 2026-grad internship drive screenshot); an Atlassian-branded timed IDE with sections was used for the 2027-grad OA (Aug 2026). UK graduate track started with psychometric tests (Sep 2026).
- Format: 2 to 4 coding questions, about 90 min (US intern 2025); India campus OLT 2 to 3 questions of roughly Codeforces 1600 difficulty (DP, graphs, bitmasking, greedy, XOR). Sydney: 2 medium plus 1 easy. Data engineering intern OAs mix SQL and LeetCode.
- Notes: OA is mandatory for early careers (official). Reported 2026 OA problems: count bitonic subsequences mod 1e9+7 (HackerRank, Q4 of 4); minimum total cost to make all array values distinct when incrementing A[i] costs C[i]. India campus shortlists are small (one 2024 campus: 15 shortlisted, 5 offers).
Prepare with How to pass an OA and compare formats in OA formats by company.
Coding rounds¶
- Rounds: Intern and new grad: 1 to 2 coding interviews (45 to 60 min). P30/P40 experienced loop: 2 coding interviews of 60 min (Data Structures + Code Design), after a Karat or technical screen.
- Style: LeetCode easy to medium for interns; medium (sometimes hard, e.g. All O`one) for P40, usually wrapped in an Atlassian product story (Jira activity, Confluence audit logs, support agent ratings, voting). Code design problems are multi-part and extensible (Snake game, tennis court booking with maintenance breaks, middleware router with wildcards, cinema scheduling, customer/product rating system). Some loops add concurrency follow-ups (P60 customer rating code design, Feb 2026).
- Environment: Your own IDE in your chosen language, screen-shared over Zoom, with your favorite test framework ready; CodePair offered if you cannot set up an IDE; using a debugger is encouraged (official backend prep PDF). AI-enabled Code Design rounds (2026) run in a HackerRank IDE with AI coding assistants and an existing repo. Frontend: browser coding in your framework (set up a page that prints the current time before the call).
- Graded on: Official: coding skills, code quality, conceptual thinking, adaptability, navigating blockers, communication; testing matters (TDD recommended); show alternative solutions and why you chose one. 'Clean code matters, but what's most important is how you think.' AI round: how you plan with AI, articulate prompts, and validate the result (candidate report Sep 2026).
- Reported focus topics: Hash maps with custom sorting and tie-breaks (voting, ratings), Extensible OOP / code design with unit tests (Snake, booking, router, rating systems), Trees: LCA in n-ary and org-chart trees, Intervals and sweep line, Design data structures (LRU, All O`one, stock price tracker, top-K), DP, graphs and bitmasking for India campus OAs, Strings and prefix sums, AI-assisted coding in an unfamiliar codebase (2026), System design (P40+): tagging, top-K feeds, crawlers, AI-era system design: agents over Jira/Confluence-style data (Project Status Agent, 2026)
Run every practice problem through the 45-minute framework and the code quality rubric.
What they ask (data)¶
Based on 5 distinct problems tagged to Atlassian in the last 6 months (1 in the last 30 days, 4 in the last 3 months, 62 all-time) across two open datasets of LeetCode company tags. Tags are user-reported, so treat frequency as a signal, not a promise.
Difficulty mix (last 6 months): Hard 40%, Medium 40%, Easy 20%
Most tagged topics (share of problems): Array 100%, Dynamic Programming 40%, Sorting 20%, Quicksort 20%, Hash Table 20%, Math 20%, Combinatorics 20%, Enumeration 20%, Number Theory 20%, String 20%
Watch out: Atlassian has thin LeetCode data. Weight the reported questions and the format notes above more than this list.
Most frequent problems¶
Ranked by frequency, weighted toward the last 30 days. Classics like Two Sum sit near the top of almost every company's list because users tag them everywhere. Solve those fast. The signature list below is more specific to this company.
| # | Problem | Difficulty | Last seen | Topics |
|---|---|---|---|---|
| 1 | Minimum Difficulty of a Job Schedule | Hard | 30 days | Array, Dynamic Programming |
| 2 | Best Time to Buy and Sell Stock | Easy | 3 months | Array, Dynamic Programming |
| 3 | Merge Intervals | Medium | 3 months | Array, Sorting, Quicksort |
| 4 | Maximum Score with Co-Prime Element | Hard | 3 months | Array, Hash Table, Math, Combinatorics |
| 5 | Word Search | Medium | 6 months | Array, String, Backtracking, Depth-First Search |
Signature problems¶
Problems where Atlassian accounts for a large share of all recent tags across companies. These are the most Atlassian-specific questions in the data.
| # | Problem | Difficulty | Last seen | Topics |
|---|---|---|---|---|
| 1 | Minimum Difficulty of a Job Schedule | Hard | 30 days | Array, Dynamic Programming |
| 2 | Maximum Score with Co-Prime Element | Hard | 3 months | Array, Hash Table, Math, Combinatorics |
Reported in 2025 to 2026 interviews¶
Questions candidates said they got, each linked to the post where it was reported.
| Question | Role | When | Source |
|---|---|---|---|
| Activity streak: longest contiguous run with constant day-to-day difference; part 2 allows replacing one value | SWE-1 (P30), India, 2027 grad | 2026-09 | post |
| Minimum total cost to make all array values distinct when incrementing A(i) by 1 costs C(i) | SWE-1 OA, India, 2027 grad | 2026-08 | post |
| Count bitonic subsequences (strictly up then strictly down, length 3+), mod 1e9+7 | SDE1 OA (HackerRank internship drive), India, 2026 grad | 2026-06 | post |
| Binary audit log with '?' wildcards: can it be filled so no palindromic substring of length 5+ exists | SDE-2 (P40) DSA round, India | 2026-08 | post |
| AI-assisted coding: fix a bug in a small backend web app (password feature) using the AI agent | Software Engineering Intern, Sydney | 2026-04 | post |
| Tennis court booking in 3 parts: assign bookings to courts, add maintenance breaks, then usage tracking | Senior SDE onsite (also P50 DSA round) | 2026-02 | post |
| Closest common organization for a list of employees (build the org tree, LCA, write tests; follow-ups add/remove users) | Senior SDE onsite coding | 2026-02 | post |
| Lowest common ancestor of a given set of n nodes in a parent-pointer tree | Intern (location not stated) | 2025-10 | post |
| Middleware router: register paths with wildcards and resolve routes (interviewer pushed a Trie) | Senior SDE (P50, leveled P40) Code Design | 2026-01 | post |
| Snake game LLD: snake grows every n-th food (or every 5 moves), game over on self-collision (LC 353 is Premium) | SDE-2 (P40) LLD round, Bangalore | 2025-09 | post |
| Rank teams by votes | SRE Intern, Sydney | 2025-03 | post |
| Support agent rating system: record 1 to 5 ratings, average, sort with tie-breaks; follow-up weighted votes | Senior Engineer coding / code design | 2026-01 | post |
| All O'one data structure | Senior Backend Engineer DS round, Bengaluru | 2026-02 | post |
| Stock price fluctuation (max price tracker with duplicate, out-of-order timestamps) | Software Engineer 2 (P40) DS round | 2025-05 | post |
| Shopping list: count department visits saved by grouping items by department | Technical phone screen (Karat style) | 2025-10 | post |
| Castle rooms graph: rooms with 2+ incoming edges that point to a room leading to treasure rooms | Karat screen, P30 (SDE1) | 2025-08 | post |
| Merge-intervals variant with sweep-line follow-ups | SWE-II (P40) DSA round | 2026-01 | post |
| LRU cache plus OOP questions | SWE Intern, New Delhi | 2025-07 | post |
| Lowest common ancestor of two values in a BST | Software Engineering Intern, Sydney | 2025-04 | post |
| Design a tagging system across Jira, Confluence and Bitbucket (system design) | Senior SDE onsite system design (Hello Interview also shows Senior, Staff and Principal reports, Nov to Dec 2025) | 2026-02 | post |
| AI-assisted coding round follow-up: why do you need both backend and frontend validation? | Software Engineering Intern, Sydney (AI-assisted round) | 2026-05 | post |
| SQL: Nth highest salary without ROW_NUMBER() or RANK(); plus a 'needle in a haystack' string-search LeetCode easy | Data Engineer Intern, Sydney | 2026-06 | post |
| Wordle feedback string, like Bulls and Cows: generateFeedback('crane', 'train') returns 'XGGXY' | P60 DSA round | 2026-02 | post |
| Closest common department finder: model a department hierarchy and find the lowest shared ancestor for a group of employees (LLD) | Mid-level (Hello Interview LLD report) | 2026-04 | post |
| Design a Project Status Agent: AI agent that ingests DevOps, Slack and meeting transcripts and answers natural-language status questions (ML system design) | Senior (Hello Interview) | 2026-08 | post |
Beyond LeetCode¶
AI-enabled Code Design (HackerRank IDE + AI assistant, extend an existing repo; one Sep 2026 report: plan and implement a Spring Boot web feature with AI and explain how you validated it). Code Design / LLD in your own IDE with JUnit-style tests (Snake game, middleware router, tennis court booking, rating systems). Karat screen with rapid-fire design scenarios. Browser coding for frontend roles (build a UI component; the official Fullstack guide asks you to pre-build a page that prints the current time). Management and values interviews. UK graduate psychometric tests.
System design¶
Not a standard round for interns or most new grads; P30 Karat screens include about 20 min of rapid-fire design scenarios (scaling bottlenecks, cost estimation, unreliable third-party API, large file uploads). From P40: one 60-min system design (whiteboard, Zoom whiteboard or drawing tool) on Atlassian-flavored prompts: tagging system across Jira/Confluence/Bitbucket, top-K popular or most-engaged Confluence pages (exclude already-viewed), web crawler; senior AI-era prompt: a Project Status Agent over DevOps, Slack and meeting data (Aug 2026). LLD is covered by the Code Design coding round (Snake, router, booking, ratings).
Start with who needs system design, then the framework.
Behavioral¶
Framework: Atlassian's five values: Open company, no bullshit; Build with heart and balance; Don't #@!% the customer; Play, as a team; Be the change you seek. Answer in STAR format. (official page)
What they look for:
- Stories mapped to each of the five values (official prep PDF lists prompts per value)
- User-centered decisions and the trade-offs you made for the customer
- Handling difficult conversations and conflict, adapting communication style
- Advocating for change and what you learned from the outcome
- Owning outcomes, using expertise to drive improvements, taking initiative, helping the team succeed (management interview areas)
- Concrete metrics for impact, since behavioral performance feeds leveling
- Values interviewer is usually from a different department than the hiring team, so keep it non-technical unless invited
Questions to prepare:
- How have you helped make a team or individual more successful if they weren't performing well?
- Have you ever had to put the team's goals above personal ambitions?
- When have you advocated to push for change? What was the outcome?
- How have you put the customer first? What trade-off did you make?
- What's your approach when having difficult conversations? How have you resolved conflict?
- What are examples when you or the team missed the mark? How did you react and fix it?
- Tell me about a time you advocated for the customer (Data Engineering Intern, 2025)
- How do you use agents to resolve issues in a repository? (SWE Intern, Kharagpur campus, Aug 2026)
- What was a trade-off you dealt with at your last internship? (SWE Intern, Vancouver, 2026)
- Give me an example of a tough or critical piece of feedback you received. (Values interview, Hello Interview, mid-level to senior manager, Jul 2026)
Build your answers with the story bank. Company detail: behavioral guide.
Tips¶
- Before any coding round, open a blank project in your strongest language with a test framework (JUnit, pytest) ready; you will write, run and test code on screen share, and a debugger is allowed.
- Drill the multi-part repeats (tennis court booking, Snake, agent ratings, router): finish part 1 fast and keep classes open for parts 2 and 3.
- For 2026 loops, practice extending an unfamiliar repo with an AI assistant and narrate how you check its output; Atlassian says it is fine to say you are new to the language.
- Write 1 to 2 STAR stories per value (5 values) with numbers; the values interviewer is often from another department, so keep stories non-technical unless asked.
- Ask clarifying questions in design rounds even when the answer seems obvious; a 2026 candidate was downleveled for skipping them.
- Apply in your region's window (US Sep to Mar, India Jul to Dec, ANZ Feb to Jun); review is rolling.
- India campus OAs are harder than the interviews (DP, graphs, bitmask around Codeforces 1600); aim to fully solve 2+ problems.
- Keep other processes going until you have a signed offer: team matching stalled for several candidates around the March 2026 layoffs.
4-week plan for Atlassian¶
Do this after you finish a core list like Grind 75 or NeetCode 150.
- Week 1: Solve problems 1 to 20 from the most frequent list. Time-box each at 30 minutes.
- Week 2: Solve problems 21 to 40. Re-solve any you failed in week 1 without looking.
- Week 3: Solve the signature problems and every reported question above. Do 2 timed mock interviews.
- Week 4: Write 8 stories for the Atlassian's five values: Open company, no bullshit; Build with heart and balance; Don't #@!% the customer; Play, as a team; Be the change you seek. Answer in STAR format. round. Do 2 full mock loops. Review the online assessment and system design notes above.
Sources¶
- Question data: liquidslr/leetcode-company-wise-problems and snehasishroy/leetcode-companywise-interview-questions, merged and ranked by scripts/build_question_data.py.
Watch out: Strong: official stages (2 coding, system design, manager, values, hiring committee), own-IDE + tests expectation, AI-enabled Code Design (official PDFs Jul/Aug 2026), values list, Levels.fyi ladder. Medium: intern/new grad loop shape varies by region (US: OA + 1 technical + EM behavioral; India campus: OLT + 1 to 2 technical + HR/values; Sydney/India 2026: AI-assisted coding); whether the AI round has replaced classic DSA for all early-career loops is unclear. OA platform for the 2027-grad India OA is not named. P30 vs 'SWE-1'/'Graduate' naming varies by region. Hiring volume after the March 2026 layoffs is uncertain; several 2026 posts describe freezes or re-evaluations. Third-party claims (300 to 500 interns/yr, 4 to 6 week timelines) are not confirmed by Atlassian. LeetCode discuss pages return 403 to scripts; their content was confirmed via LeetCode's GraphQL API. Glassdoor review permalinks were confirmed in a browser session (they block non-browser clients). Web search budget ran out mid-task, so Reddit and Blind were not covered. FACT-CHECK (2026-10-04, second agent): all 90+ URLs re-fetched (atlassian.com pages and PDFs via curl; LeetCode problems and discuss posts via LeetCode GraphQL incl. OA screenshots viewed; Glassdoor permalinks and Hello Interview timelines rendered in a browser). Fixed: several Glassdoor reviews do not state a location (removed 'US', 'Australia', 'IIT' labels); tagging-system SD had no 2026-01 or mid-level report on Hello Interview, so it now cites the Feb 2026 onsite post; P40 timeline post does not say India; concurrency claim now cites the P60 report; the 'Sydney OA 2 mediums and 1 easy' claim could not be found and was removed. Premium LeetCode links: LC 353 Design Snake Game. Removed unverifiable specifics (frontend 'tic tac toe' and 'nested bullet list', SD 'scorecard service' and 'color picker', code design 'rate limiter' as an asked question): not found in Glassdoor pages 1 to 22, the cited LeetCode posts, or the official pages; the official Frontend and SRE guides are Google Drive files that were not opened.
Next: All companies