Decomposition and beyond: frontend engineering at scale
summary
Atlassian describes how Jira’s frontend grows to more than one million lines with 200+ developers across five time zones and over 50 apps, and how that scale forces concrete choices: adopt a tangerine architecture of layered, folder-mirrored apps; automate with generators, ESLint and Prettier; add a project-wide static analyzer stricter with a circular-deps rule; enforce ownership via manifest files and a golem bot on pull requests. The result is reproducible structure and failing builds when rules break.
AI-written summary · about
chapters (10)
- 0:00Start
- 2:53TECH STACK
- 3:07JIRA FRONTEND
- 7:14LEARNING 1 DIVIDE AND CONQUER
- 7:56TANGERINE SEPARATION OF CONCERNS
- 9:00VIEWS STRUCTURE
- 12:21STATE STRUCTURE
- 13:21TANGERINE STATE + VIEW
- 17:57RADICALLY AUTOMATE STRICTER
- 20:09LESSON 3 OWNERSHIP
related talks
25mLessons for frontend development at scale
18mBeyond the Hype: Facing Micro Frontend Challenges in Large-Scale
25mThe tragedy of scale
26mBeyond the Hype: Facing Micro Frontend Challenges in Large Scale
58m’s journey to cloud-native architecture
40mBuilding resilient frontend architecture
58mDeveloper Joy – How great teams get s%*t done
35mMicro frontend Has Entered the Game