Netflix describes how the Localization team modernized its analytics infrastructure to support 300M+ members across 190+ countries and 50+ languages.
This post details Netflix's multi-step CPU optimization journey for video serendipity scoring in the Ranker recommendation service using JDK's Vector API.
Netflix explains how migrating to kubelet + containerd with per-container user namespaces triggered severe mount lock contention on r5.metal nodes.
Netflix introduces MediaFM, an in-house tri-modal (audio, video, text) foundation model for deep media content understanding at scale.
This article describes Netflix's internal Post-Training Framework for adapting LLMs to production recommendation, personalization, and search use cases.
Netflix's Online Data Stores team built an automated self-service workflow to migrate approximately 400 RDS PostgreSQL clusters to Amazon Aurora PostgreSQL.
Netflix evolved its Graph Search platform to support natural language queries using LLMs, translating user input into their internal Graph Search Filter DSL.
This post explains how Netflix adopted Temporal, a durable execution platform, to improve reliability of cloud operations in their Spinnaker CD pipeline.
This post describes the Netflix Live Origin server, a custom microservice that brokers live streaming content between cloud encoding pipelines and the Open Connect CDN.
This post details Netflix's AV1 codec adoption journey, which now powers approximately 30% of all Netflix streaming.
