High-Level Design (system design) rounds at LinkedIn lean toward real internal-style problems — notification fan-out, key-value stores, ranking/aggregation pipelines — rather than pure "design Twitter" templates. Interviewers reward a structured approach: requirements first, then scale estimation, core entities, architecture, and a deep dive into one component.
Problem 1Design LinkedIn's Notification System
Design a system that delivers notifications (e.g. connection requests, post engagement, messages) to users at scale across push, email and in-app channels.
Requirements to clarify first
- Which events are one-to-one (a connection request) and which can fan out to a huge audience (a popular post being liked, a company page posting) — the second kind needs a different write strategy?
- How fresh must a notification be — is a few seconds of delay acceptable for a like, while a message needs near-instant push?
- Can a user turn channels on and off per notification type, and must a notification read on one device show as read on every other device and in email?
More system design prompts LinkedIn has asked recently
The full architecture, API design, data model and scaling discussion for these are part of the Placement-Ready PYQ Kit.
2 more system design prompts with a full architecture writeup in the full Placement-Ready PYQ Kit