How do I approach ML System Design interview questions?

ML System Design questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master ml system design interviews.

What difficulty level is this interview question?

This is a medium difficulty ML System Design question, commonly asked during Onsite rounds at Figma.

What role is this question designed for?

This question is commonly asked for Machine Learning Engineer candidates at Figma during technical interviews.

Design Feed Ranking and HTML Generation

Last updated: Apr 16, 2026

Quick Overview

This question evaluates the ability to design end-to-end machine learning systems for personalized feed ranking and for natural-language-to-HTML generation, testing competencies in objectives setting, candidate generation and ranking, feature engineering, cold-start handling, exploration, online serving, offline and online evaluation, data collection, model architecture, supervised fine-tuning, reinforcement learning or preference optimization, safety constraints, and the use of LLMs for judging. It is commonly asked to probe architectural trade-offs, scalability and production operational considerations, falls under the ML System Design and applied machine learning domain, and assesses both conceptual understanding of high-level design and practical application knowledge for production implementation.

Figma

Mar 25, 2026, 12:00 AM

Machine Learning Engineer

Onsite

ML System Design

The ML system design portion covered two prompts.

Design a personalized home feed for a collaborative design product. Explain the end-to-end ML system, including objectives, candidate generation, ranking, features, cold start handling, exploration, online serving, and offline and online evaluation.
Design a system that takes a natural-language description of a webpage and generates the corresponding HTML. Discuss data collection, model architecture, supervised fine-tuning, reinforcement learning or preference optimization, safety constraints, serving, and evaluation. Also discuss whether and how to use an LLM as a judge.

Solution

Show

Submit Your Answer to Earn 20XP

Loading comments...

Browse More Questions

More ML System Design•More Figma•More Machine Learning Engineer•Figma Machine Learning Engineer•Figma ML System Design•Machine Learning Engineer ML System Design