linkedin | 全职 | Interview Experience
Interview Date: Not specified
Result: Not specified
Difficulty: Not specified
Interview Process
The interview process consisted of four rounds:
-
AI Coding: Implemented an LRU Cache, followed by a multi-threaded version of the LRU. The AI’s role was to search for infrequently used APIs, which made the testing phase quite tricky.
-
System Design: Designed a metrics collection and querying system that collects metrics from production systems and supports queries of arbitrary granularity, with results sent to an external visualization system. The problem was somewhat vague, and the interviewer changed requirements and design during the discussion, leading to a less engaging interaction.
-
Behavioral Interview: Questions focused on motivation for joining LinkedIn, current company level and rating, attempts at promotion, and outcomes of those attempts. The interviewer frequently interrupted to signal understanding but later asked for details, indicating a lack of clarity on her part.
-
Coding: Implemented an
AlertMonitorclass with three functions:- Summarize the number of alerts within the last 15 minutes given a timestamp.
- Summarize the distribution of alerts given a timestamp.
- Calculate alert spikes, defined as the index of the first time window where the alert count is lower than the current count. Significant time was spent understanding the problem, and there was not enough time to discuss optimizations.
Technical Questions
- LRU Cache (Hash Table, Doubly Linked List, Design)
- System Design (Time-Series Database, Scalability)
- Behavioral Interview Questions (Motivation, Career Development, Problem Solving)
Tips & Insights
- Be prepared for vague questions in system design and be proactive in seeking clarification.
- Expect behavioral questions to focus on your motivations and career progression.
- Time management is crucial during coding interviews, especially when discussing optimizations.