Video Recommendation: Problem Framing and Requirements
Explore how to properly frame a video recommendation system problem by asking focused clarifying questions on product, user, and content context. Understand how to translate business objectives into measurable ML metrics and factor scale constraints. This lesson prepares you to define clear problem statements that guide effective design decisions in ML system interviews.
In an ML system design interview, you may be asked: “Design a video recommendation system for a large-scale video platform.” A strong answer starts by framing the problem before jumping into architecture. Instead of listing components without a clear structure, build the design around requirements, constraints, and trade-offs. Video recommendation is a common ML system design interview prompt because it brings together several core production ML challenges: candidate generation over large catalogs, real-time ranking under strict latency budgets, feedback loops that can reinforce poor recommendations or stale preferences, and business objectives that can conflict, such as engagement, satisfaction, freshness, and revenue.
The dominant paradigm for these systems is a two-stage architecture where a fast candidate generation layer narrows billions of videos down to hundreds, and a heavier ranking layer scores those candidates for the final recommendation slate. This lesson is not about architecture yet. It covers the work that comes before drawing the architecture. Problem framing and requirements gathering are upstream work that shape the main design decisions, and skipping them is a common failure mode in ML system design interviews. Many candidates rush into modeling. A stronger answer pauses here to clarify the problem first.
Clarifying questions for video recommendation
A well-structured set of clarifying questions signals to the interviewer that you understand the problem space deeply enough to know where ambiguity hides. Rather than asking generic questions, each question should directly influence an architectural or modeling decision downstream.
Organizing questions by category
The following categories cover the critical dimensions of a video recommendation problem:
Product context: Asking whether the recommendation surface is a home page feed, a watch-next sidebar, or a search results page determines whether candidate generation must cast a wide net across diverse topics or stay contextually anchored to the video currently playing.
User context: Understanding ...