Netflix
4 articles in the last 30 days.
-
QCon San Francisco 2026: Scaling Tech by Aligning Humans
Netflix's Scott Behrens presents a playbook for staff-plus engineers to find a problem, pitch it, form a small tiger team and set a north star. The aim is a company-wide technical strategy built from clear bets.
-
Engineering AI Discover
Netflix describes the agent harness behind AI Discover, a TV tab for steerable discovery. It combines LLMs with recommenders under seven design challenges, including latency, grounding, and safety.
-
Fact-Checking at Scale: Building a More Accurate and Efficient Synopsis Factuality System
Netflix replaces an expensive reasoning-model factuality checker for synopses with a cheaper system: 52.4% lower cost per synopsis, Recall-Fail up from 0.77 to 0.86, and better rationale alignment.
-
How Netflix Uses AI Agents for Performance Engineering
Describes Netflix's autonomous performance agent that automates profiling and bottleneck analysis, and the problems it found in production, including the safety measures it needed.