Netflix

In-House LLM Serving at Netflix

Netflix needed to deploy and serve large language models at scale within their production environment rather than relying on third-party hosted APIs, while maintaining the reliability and performance standards required for their streaming platform.

microservices ml-systems
5 min
Netflix

GenPage: Towards End-to-End Generative Homepage Construction at Netflix

Netflix needed to generate personalized homepages at scale where every row, entity ordering, and layout element is customized per user while maintaining low latency and high relevance.

ml-systems real-time-systems
5 min
Netflix

A Human-Augmenting Agentic Workflow for Causal Inference

Building a data analysis system that can reliably infer causal relationships from observational data while accounting for hidden biases and confounding variables that automated agents might miss.

ml-systems observability
5 min
Netflix

Predicting Risk in Content Launches: How Data-Driven Insights can Transform Launch Planning

Netflix needed to predict and mitigate risks associated with content launches to improve availability and reduce failures during high-impact release events.

observability microservices
5 min
Netflix

Thinking Fast & Slow for a Personalized Notification System

How to design a notification system that intelligently decides when and how to deliver personalized notifications to users with varying urgency and cognitive load requirements.

microservices real-time-systems
5 min
Netflix

Toward More Controllable AI Video Editing: An Early Research Exploration at Netflix

Netflix needed to enable video editors to have more fine-grained control and predictability over AI-assisted video editing systems for creating promotional content at scale.

ml-systems api-design
5 min
Netflix

VMAF v1: Good Is Not Good Enough

Netflix needed to improve the accuracy of video quality assessment metrics beyond VMAF v0 to make better encoding decisions and prevent quality misjudgments that could degrade the member experience.

ml-systems observability
5 min
Netflix

A Human-Augmenting Agentic Workflow for Causal Inference

Building a reliable causal inference system that can accurately determine cause-and-effect relationships in data (like the impact of a show on retention) while accounting for hidden biases and confounding variables that automated agents might miss.

ml-systems observability
5 min
Netflix

Predicting Risk in Content Launches: How Data-Driven Insights can Transform Launch Planning

Netflix needed to predict and mitigate risks associated with content launches to improve reliability and reduce unexpected failures in production.

observability ml-systems
5 min
Netflix

Thinking Fast & Slow for a Personalized Notification System

Netflix needed to design a personalized notification system that efficiently decides when and how to notify users with relevant content recommendations without overwhelming them or missing critical engagement opportunities.

microservices ml-systems
5 min
Netflix

VMAF v1: Good Is Not Good Enough

Accurately measuring video quality perception to optimize encoding decisions and ensure Netflix members receive the best possible streaming experience across different bitrates and codecs.

ml-systems observability
5 min
Netflix

Democratizing Machine Learning at Netflix: Building the Model Lifecycle Graph

Netflix needed to manage the lifecycle of machine learning models across multiple domains and teams at scale, moving beyond their original single-domain personalization focus.

ml-systems microservices
5 min
Netflix

Evaluating Netflix Show Synopses with LLM-as-a-Judge

Netflix needed to automatically evaluate the quality and relevance of show synopses at scale to improve member discovery and engagement.

ml-systems api-design
5 min
Netflix

Powering Multimodal Intelligence for Video Search

Netflix needed to efficiently extract and surface key moments from hundreds or thousands of hours of raw video footage for editorial teams to accelerate the creative content production process.

ml-systems search
5 min
Netflix

State of Routing in Model Serving

Netflix needed to design a domain-independent traffic routing system for their ML model serving infrastructure that could handle personalized experiences at scale across multiple domains while maintaining high availability.

microservices load-balancing
5 min
Netflix

MediaFM: The Multimodal AI Foundation for Media Understanding at Netflix

Netflix needed scalable, deep machine-level understanding of every piece of content across an expanding catalog (including live events and podcasts) to power recommendations and discovery, but building separate models per content type and modality doesn't scale.

ml-systems microservices
5 min
Netflix

Optimizing Recommendation Systems with JDK’s Vector API

Netflix's Ranker service had a video serendipity scoring feature (computing how different a title is from a user's watch history) consuming ~7.5% of total CPU per node, creating a significant performance bottleneck at their enormous scale.

ml-systems real-time-systems
5 min
Netflix

Scaling LLM Post-Training at Netflix

Generic pre-trained LLMs lack the domain-specific alignment needed for Netflix's production use cases in recommendation, personalization, and search, and the post-training pipeline to fine-tune them doesn't scale efficiently across multiple domain constraints and reliability requirements.

ml-systems distributed-systems
5 min
Netflix

The AI Evolution of Graph Search at Netflix

Netflix's Graph Search platform for federated enterprise data required users to write structured queries, limiting accessibility and ease of use despite the system being scalable and configurable.

search ml-systems
5 min