🔸Netflix ML Architecture🔸

Hey Guys, Have you wondered how netflix recommends you the videos you might ne intrested to watch.

Let me show you have it workss

Stayy tunned ..............

👉In Netflix videos are recommended based on members interests using ML algorithms

👉Here let us have a look at the important components of Netflix ML Architecture.

👉Axion Fact Store:
Axion fact store is part of Netflix Machine Learning Platform, the platform that serves machine learning needs across Netflix. First part of below image shows how Axion interacts with Netflix’s ML platform.

👉Fact:
A fact is data about the members or videos. An example of data about members is the video they had watched or added to their My List. An example of video data is video metadata, like the length of a video.

👉Compute application:
These applications generate recommendations for the members. They fetch facts from respective data services, run feature encoders to generate features and score the ML models to eventually generate recommendations.

👉Offline feature generator:
Offline Feature Generator is a spark application that enables on-demand generation of features using new, existing, or updated feature encoders.

👉Shared feature encoders:
Feature encoders are shared between compute applications and offline feature generators.

👉ETL and Query Client:
ETL and Query Client are intertwined, as any ETL changes could directly impact the query performance. ETL is the component where the experiment for query performance, improving data quality, and storage optimization are done.

👉 Note : The second part of below image shows components of Axion’s ETL and its interaction with the query client.


source : Netflix Technology Blog

Share if you like #analystdhiren