10 posts · searchable
10 posts
Netflix's JVM Ecosystem team built Nebula ArchRules to scale ArchUnit for enforcing architecture rules across tens of thousands of Java repositories.
Netflix describes building the Model Lifecycle Graph to democratize machine learning across the company.
This post introduces a series on ML model serving infrastructure, focusing on routing strategies for personalized experiences at scale.
Netflix's Content Hub team scaled camera file processing through strategic collaboration and a media production suite.
Netflix details the human and operational infrastructure built to support live streaming at scale over three years.
Netflix uses LLM-as-a-judge to evaluate and improve show synopses for personalized member recommendations.
Netflix implemented interval-aware caching for Apache Druid to avoid redundant queries at massive scale.
Netflix built a multimodal intelligence system to enable video search across hundreds of hours of raw footage.
Netflix rolled out variable bitrate encoding for all live events, improving streaming efficiency and quality.
Netflix modernized its localization analytics to support global storytelling and content scaling.