System design (HLD) rounds at Atlassian favor grounded, product-shaped prompts — an internal tags system, a per-user preferences store — over abstract "design Twitter" templates, and interviewers push specifically on where your design would bottleneck first as scale increases.
Problem 1Design an Internal Tags-Management System
Design an internal system that lets teams create, look up, and reuse tags across an Atlassian-scale product. The interviewer will push on where your design would bottleneck as tag creation and lookup volume grows.
Requirements to clarify first
- Are tags scoped per tenant/site (two customers can each have their own
urgenttag) or global across the product — and can one tag be attached to several entity types (issues, pages, pull requests)? - What is the read/write mix — is lookup (autocomplete while typing, listing tags on an entity) 100x more frequent than tag creation, and how fresh must a newly created tag be in search results?
- Are tags free-form strings that anyone can create, or is there a curated vocabulary with rename/merge/deprecate operations that must propagate to every tagged entity?
More system design prompts Atlassian has asked recently
The full architecture, API design, data model and scaling discussion for this is part of the Placement-Ready PYQ Kit.
1 more system design prompt with a full architecture writeup in the full Placement-Ready PYQ Kit