Netflix Shares LLM-as-a-Judge Production Practices at Scale

Netflix published a paper detailing how it runs LLM-as-judge weekly on hundreds of thousands of recommendation explanations serving millions of mobile users, treating the judge as a lifecycle-managed agent to keep it reliable in production.

2026-08-24 ~ 2026-08-24 · 3 related posts

• Netflix details its production LLM judge: hundreds of thousands of recommendations scored weekly — omarsar0 · 2026-08-24
• Netflix paper details the full lifecycle of LLM-as-a-Judge in production recommenders — tokenbender · 2026-08-24
• Netflix details LLM-as-judge design for recommender systems — dejavucoder · 2026-08-24