← LinkedIn Interview Insights

LinkedIn·Data Scientist·Onsite - System Design / Architecture·Senior

Senior
May 2026

Summary

LinkedIn data scientist interview with a meaty system design question on short-video recommendations. The scope was pretty wide, covering everything from labels and features to offline evaluation and online guardrails, so you really had to show both ML chops and product thinking at the same time.

Questions Asked (1)

Q1

Design a recommendation system for a short-video feed at scale, covering the full pipeline from objectives and labels through to online evaluation and guardrails.

System DesignProduct Analytics & MetricsTechnical Trade-offs
Author's notes

This question is basically six questions stitched together and the interviewer will pull on whichever thread they want, so pacing yourself is genuinely hard.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Start by clarifying the product objectives and success metrics, then walk through the full pipeline: data collection, candidate generation, ranking, and online evaluation. Emphasize trade-offs between relevance, diversity, and freshness, and how guardrails protect user experience and platform health.

Pro tip: Frame the system around LinkedIn's professional context: short videos should drive meaningful engagement (e.g., skill-building, networking) not just watch time. Show you understand that optimizing for long-term value requires careful label design and guardrails against clickbait.

1. Define Objectives and Metrics

Clarify business goals (e.g., increase meaningful engagement, retention) and translate them into measurable online metrics (e.g., watch time, likes, shares, comments) and offline proxies. Consider counter-metrics like report rate or skip rate.

2. Data and Label Design

Identify data sources (user interactions, video metadata, social graph) and define labels for training (e.g., binary engagement, watch percentage, dwell time). Address biases like position bias and feedback loops.

3. Modeling and Ranking Pipeline

Outline a two-stage architecture: candidate generation (e.g., collaborative filtering, content-based) and ranking (e.g., deep learning models). Discuss feature engineering, model choices, and trade-offs between relevance, diversity, and freshness.

4. Online Evaluation and Experimentation

Describe A/B testing methodology, metric selection, and how to measure long-term effects. Include interleaving or bandit approaches for faster iteration.

5. Guardrails and Monitoring

Define guardrail metrics (e.g., user reports, unfollows, session abandonment) and monitoring systems to detect degradation. Discuss fallback strategies and ethical considerations.

Key Points to Mention

  • Two-stage architecture: candidate generation and ranking to handle scale
  • Label design: using watch time, likes, shares, and negative signals (skips, reports)
  • Handling position bias and feedback loops in training data
  • Trade-offs: relevance vs. diversity vs. freshness; short-term vs. long-term engagement
  • Online evaluation: A/B testing, interleaving, and guardrail metrics
  • Scalability considerations: distributed training, real-time inference, and feature stores

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.