Design Challenge: Personalized Recommendations for a 50-Million-Item Catalog

Difficulty: Hard · Concept: The 9-Step Framework

Same no-auto-grading, self-assessment format as the RAG design challenge — write a real end-to-end design in free text, then reveal a rubric and the site's own real walkthrough for this exact problem.

The problem: a fourth shape. You've been asked to design a system that ranks the right handful of items out of 50 million, differently for every single user, where the system's own past recommendations quietly shape the data it gets to learn from next — a feedback loop none of the other three case studies on this site have to contend with at all.

Design it end to end — how do you go from 50 million items down to a personalized, ranked handful fast enough for every request, what happens for a brand-new user or item with no history yet, and how does the model avoid quietly reinforcing its own past mistakes?

System design challenges complete for now. Back to Practice Problems for the rest of the catalog.