Reading room.
Engineering blogs & AI research
A reading room from engineering teams and AI labs: titles, short feed excerpts, and links to the originals. For my own essays, explore the writing archive.
- 1359 articles tracked
- 37 sources
- Refresh scheduled daily
Sources
Fact-Checking at Scale: Building a More Accurate and Efficient Synopsis Factuality System
By Shreya S Ramanujam , Yongce Li , Afshine Amidi , Sina Ghiassian , Benjamin Klein . Introduction Picture this — you’re home after a long day at work and can’t wait to catch up on Bridgerton. You log on to Netflix, sear…
Making Failure Diagnosis Reliable with Pensive AI
By Praneeth Yenugutala , Victor Dusa , Anastasia Samoilova , Megan Ren , Justin Wei , Marcelo Mayworm At Netflix, data and machine learning practitioners run millions of workloads each day across orchestration systems an…
Multi‑Ratio Video Resizing for Mobile Feed and Marketing
Serdar Cellat , Yuchen Xie , Bernadette Marcu , Blaid Burgess , Cody Jarrett , Carlos Pereira , Qing Shi , Samuel Carvajal , Manuel Pamplona , Dorothy Jung , Feiko Ritsema , Benjamin Klein , Alex Woo , Eugene Lok , Anna…
Beyond API Compatibility: Building a Tool to Measure the Blast Radius of Library Changes
By Roberto Perez Alcolea , Java Platform Team Quasar connects a proposed API change with observed references in consumer repositories, giving the producer evidence for a release decision At Netflix, shared Java libraries…
Single-tenant Titus: VM-level isolation on a multi-tenant container platform
Samartha Chandrashekar Nick Parker You click play on a Friday night, and in the second that follows, dozens of services answer in milliseconds. Most of them run on Titus, our internal container platform. Titus is, in ess…
Ontology-Driven Observability: Building an End-to-End Knowledge Graph at Netflix Scale
Prasanna Vijayanathan , Renzo Sanchez-Silva When Jake Paul and Mike Tyson stepped into the ring in November 2024, we delivered the fight to roughly 65 million concurrent streams. On an ordinary day our infrastructure tak…
Behind the Scenes: Evolving Netflix’s Ads Event Pipeline for Live — Part II
By Yogesh Nagarur Introduction In Part 1 of this series , we shared how Netflix built the Ads Event pipeline and approached it as a system design problem: replacing fragmented pipelines with a centralized collection, enr…
DataJunction as Netflix’s answer to the missing piece of the modern data stack
by Olek Gorajek , Samuel Redai , Yian Shang Intro Our Netflix journey from data to dashboards, particularly within the space of the experimentation platform, always presented a unique set of challenges. This blog post ai…
Integrating Netflix’s Foundation Model into Personalization applications
Authors: Divya Gadde , Ko-Jen Hsiao , Dhaval Patel and Moumita Bhattacharya The motivation for building the Netflix Foundation Model for personalization was laid out in this previous blog post. The Netflix homepage is po…
Measuring the Impact of Personalized Recommendations
By Kevin Zielnicki, Guy Aridor, Aurélien Bibaut, Allen Tran, Winston Chou, and Nathan Kallus Introduction Personalization is everywhere on the internet. It shapes the products we discover, the stories we read, the music…
Netflix Conductor: The Next Chapter
Netflix Conductor : The Next Chapter by Aravindan Ramkumar on behalf of the Conductor team Conductor is the workflow orchestration engine Netflix uses to stitch micro-services into reliable, observable business processes…
Netflix Stratum Media Processing: Automated Container Right-Sizing
By Violetta Pidvolotska and Naveen Mareddy Stratum is Netflix’s internal serverless media processing platform, shaped by over a decade of experience developing large-scale distributed systems. It powers a wide range of m…
Reliable GPU compute for large training & inference jobs
How we detect failing GPUs and network fabric, remediate machines automatically, and let long-running jobs survive hardware loss. Authors: Nick Rutigliano , Argha C , Samartha Chandrashekar Open Netflix, and the home scr…
Running Apache Spark experiments in my sleep (and on a plane)
By Prashanth Gedde Narayanaswamy As one of Netflix’s largest and critical data pipelines, member-sessionizer consolidates tens of billions of hourly client events, encompassing user taps, playbacks, and engagement signal…
The Lifecycle of LLM-as-a-Judge: Building, Aligning, and Monitoring at scale
By Emma Yanyang Kong , JJ Tan , Ishan Gupta , Lars Olds , Claire Campbell , David Fagnan , Ratna Kavuri , Rohan Gosain , Veli Balin , Minsu Jang , Louis Garcia Introduction Any team generating text with an LLM at scale h…
TimeSeries Tiered Storage Journey: Kafka/Flink Streams to Native Cassandra Cold Reads
By Oleksii Tkachuk , Rajiv Shringi , Chris Lohfink , Abe Ratnofsky , Kartik Sathyanarayanan Introduction In an earlier post , we introduced the TimeSeries Abstraction, our high-throughput, low-latency platform designed t…
Trading a Cloud Identity for Your Own: Workload Attestation on Managed Compute
By Dhruv Pratap Introduction Organizations that have been around for a while usually run two identity systems side by side. One belongs to the cloud provider: IAM roles, instance profiles, execution roles. The other is y…
Leave the Class Path in the Rearview Mirror
Introducing composable, module system native and agent friendly command line tools for modern Java development By Danny Thomas, JVM Ecosystem Team Recent work on the Java language to pave the on-ramp has made it easier t…
MAPS: Netflix’s Multimodal Asset Personalization at Scale
By Emma Yanyang Kong , Aditya Deshpande , Asad Abbasi , Bowei Yan , David Fagnan , Ashish Rastogi , Dhaval Patel , Ray Zhang Introduction The Netflix experience is a journey of discovery. Every visual cue, from the artwo…
A Tale of Two Flink Autoscalers
Samuel Yeboah , Francesco Di Chiara and Mingliang Liu Today, Netflix runs two Flink autoscalers. That is exactly one more than we want. We built the first one in-house years ago, when there was no mature option suited to…
How and Why Netflix Built a Real-Time Distributed Graph: Part 3 — Querying the graph with gRPC…
How and Why Netflix Built a Real-Time Distributed Graph: Part 3 — Querying the graph with gRPC execution API Authors: Nilesh Mishra and Ajit Koti This is the third entry of a multi-part blog series describing how we buil…
Modeling Device Capabilities for Analytics
by Aarti Laddha , Richard Diaz-Cool , Rishika Idnani , Venkatesh Selveraj Netflix supports a vast and evolving set of features and content types, ranging from 4K streaming and immersive audio to live streaming and cloud…
GenRec: Towards LLM-Native Recommendation at Netflix
Authors: Ying Li , Arjun Rao , Shradha Sehgal Introduction Recommendations sit at the heart of the Netflix experience. Our current production models rely on thousands of hand‑crafted features over users, items, and inter…
In-House LLM Serving at Netflix
By AI Platform’s Model Runtime team and Inference team Introduction Most organizations consume LLMs through hosted APIs. Netflix went further — we run the full stack ourselves, from model deployment through inference, in…
Building Service Topology at Scale: Architecture, Challenges, and Lessons Learned
By Parth Jain , Rakesh Sukumar , Yingwu Zhao , Renzo Sanchez-Silva Nathan Fisher A deep dive into the engineering challenges of building a real-time service dependency map at Netflix scale: from streaming architectures a…
GenPage: Towards End-to-End Generative Homepage Construction at Netflix
Authors: Lequn Wang , J iangwei Pan , and Linas Baltrunas Figure 1. Autoregressive homepage generation. GenPage builds a Netflix homepage one row or entity at a time, each one conditioned on what’s already on the page an…
Toward More Controllable AI Video Editing: An Early Research Exploration at Netflix
By Zhuoning Yuan , Ta-Ying Cheng , Benjamin Klein , Bahareh Azarnoush Introduction At Netflix, we build technology to help storytellers bring their creative visions to life and to help members discover the stories they l…
How Netflix Simplified Batch Compute with Kueue
By Alvin Bao , Alex Petrov , Jennifer Lai , Aidan Sherr , and Samartha Chandrashekar As a part of the journey to transition Netflix’s compute infrastructure to be more Kubernetes-native, we have leaned into incorporating…
A Human-Augmenting Agentic Workflow for Causal Inference
By Winston Chou, Adrien Alexandre, Lars Olds, Yi Zhang, Garrett Hagemann, and Nathan Kallus Introduction Imagine asking a data agent to analyze the causal relationship between two variables, such as the effect of watchin…
Data Projects: Managing Data Assets at Netflix Scale
By Amer Hesson , Marcelo Mayworm , James Mulcahy , and Brittany Truong The Problem: Managing Assets at Netflix Scale Netflix’s Data Platform is vast. We have millions of tables in our data warehouse and tens of thousands…
From Silos to Service Topology: Why Netflix Built a Real-Time Service Map
By Parth Jain , Rakesh Sukumar , Yingwu Zhao , Renzo Sanchez Nathan Fisher How we built a living map of our distributed infrastructure to help engineers understand dependencies, troubleshoot faster, and keep Netflix runn…
Predicting Risk in Content Launches: How Data-Driven Insights can Transform Launch Planning
by Emily Gill Each year, we bring the Analytics Engineering community together for an Analytics Summit — a multi-day internal conference to share analytical deliverables across Netflix, discuss analytic practice, and bui…
The Data Canary: How Netflix Validates Catalog Metadata
By Celina Amados At Netflix, our catalog metadata is crucial to our member experience, and a single corrupted data state can impact millions of viewers immediately. To protect streaming reliability, we built an automated…
The Evolution of Cassandra Data Movement at Netflix
By Guil Pires , Jennifer Prince , Jose Camacho , Ken Kurzweil , Phanindra Chunduru Background In a previous post, we introduced Data Bridge , a unified management plane for batch Data Movement at Netflix. Historically, s…
Thinking Fast & Slow for a Personalized Notification System
by Matthew Wood , Ishan Gupta , Kevin Mercurio, Devon Bryant , and Claire Dorman In his seminal book “Thinking, Fast and Slow,” Daniel Kahneman describes two systems that drive human cognition: System 1, which operates a…
VMAF v1: Good Is Not Good Enough
By Christos G. Bampis , Zhi Li , Kyle Swanson, Nil Fons Miret and Pavan Madhusudanarao Will this encode look good to Netflix members? Does switching to a new codec improve quality at the same bitrate and by how much? Wha…
A Human-Augmenting Agentic Workflow for Causal Inference
By Winston Chou, Adrien Alexandre, Lars Olds, Yi Zhang, Garrett Hagemann, and Nathan Kallus Introduction Imagine asking a data agent to analyze the causal relationship between two variables, such as the effect of watchin…
Data Projects: Managing Data Assets at Netflix Scale
By Amer Hesson , Marcelo Mayworm , James Mulcahy , and Brittany Truong The Problem: Managing Assets at Netflix Scale Netflix’s Data Platform is vast. We have millions of tables in our data warehouse and tens of thousands…
From Silos to Service Topology: Why Netflix Built a Real-Time Service Map
By Parth Jain , Rakesh Sukumar , Yingwu Zhao , Renzo Sanchez-Silva Nathan Fisher How we built a living map of our distributed infrastructure to help engineers understand dependencies, troubleshoot faster, and keep Netfli…
Predicting Risk in Content Launches: How Data-Driven Insights can Transform Launch Planning
by Emily Gill Each year, we bring the Analytics Engineering community together for an Analytics Summit — a multi-day internal conference to share analytical deliverables across Netflix, discuss analytic practice, and bui…
The Data Canary: How Netflix Validates Catalog Metadata
By Celina Amados At Netflix, our catalog metadata is crucial to our member experience, and a single corrupted data state can impact millions of viewers immediately. To protect streaming reliability, we built an automated…
The Evolution of Cassandra Data Movement at Netflix
By Guil Pires , Jennifer Prince , Jose Camacho , Ken Kurzweil , Phanindra Chunduru Background In a previous post, we introduced Data Bridge , a unified management plane for batch Data Movement at Netflix. Historically, s…
Thinking Fast & Slow for a Personalized Notification System
by Matthew Wood , Ishan Gupta , Kevin Mercurio, Devon Bryant , and Claire Dorman In his seminal book “Thinking, Fast and Slow,” Daniel Kahneman describes two systems that drive human cognition: System 1, which operates a…
VMAF v1: Good Is Not Good Enough
By Christos G. Bampis , Zhi Li , Kyle Swanson, Nil Fons Miret and Pavan Madhusudanarao Will this encode look good to Netflix members? Does switching to a new codec improve quality at the same bitrate and by how much? Wha…
Dynamic Repartitioning for Time Series Workloads
By Rajiv Shringi , Kaidan Fullerton , Oleksii Tkachuk and Kartik Sathyanarayanan Introduction Netflix’s TimeSeries Abstraction is a scalable system for ingesting and querying petabytes of temporal event data with millise…
Dynamic Repartitioning for Time Series Workloads
By Rajiv Shringi , Kaidan Fullerton , Oleksii Tkachuk and Kartik Sathyanarayanan Introduction Netflix’s TimeSeries Abstraction is a scalable system for ingesting and querying petabytes of temporal event data with millise…
High-Throughput Graph Abstraction at Netflix: Part I
By Oleksii Tkachuk , Kartik Sathyanarayanan , Rajiv Shringi Introduction Netflix has a diverse range of graph use cases, each serving specific business needs with unique functionality and performance requirements. These…
High-Throughput Graph Abstraction at Netflix: Part I
By Oleksii Tkachuk , Kartik Sathyanarayanan , Rajiv Shringi Introduction Netflix has a diverse range of graph use cases, each serving specific business needs with unique functionality and performance requirements. These…
Scaling ArchUnit with Nebula ArchRules
By John Burns and Emily Yuan Introduction At Netflix, we operate using a polyrepo strategy with tens of thousands of Java repositories. This means that we need to have ways of sharing common build logic across these repo…
Scaling ArchUnit with Nebula ArchRules
By John Burns and Emily Yuan Introduction At Netflix, we operate using a polyrepo strategy with tens of thousands of Java repositories. This means that we need to have ways of sharing common build logic across these repo…