Walkthrough
How Netflix Recommendations Work
Step-by-step from user watch history to personalized picks
User Watches Content
A user watches movies/shows on Netflix. Every interaction (watch, pause, skip, rate) is recorded as a signal. These signals build the user's taste profile over time.
Log
Algorithm
DP (0/1 Knapsack) selects optimal movie set.
DP[i][w] = max score with first i movies, w selections. Score = genre pref + rating + recency.
Complexity
Interview Qs
Design Netflix recommendation for 200M+ users.
Collaborative + content-based filtering. Matrix factorization.
Compute user similarity at scale.
Approximate nearest neighbor (ANN). LSH for dimensionality reduction.
Movie rating prediction with matrix factorization.
Decompose into latent factors. SGD to minimize RMSE.
Handle cold-start for new users/movies.
Content-based using genre/actors. Hybrid approach with metadata.
Coding Challenge
Practice on Terminal
Write and submit your solution, then check achievements
Netflix wants to build a movie recommendation engine using collaborative filtering. Given an N×M matrix where rows are users and columns are movies (ratings 1-5, 0 = unrated), predict the missing rating for a target user and movie using the K most similar users' ratings weighted by similarity.