Netflix tests GenRec on 10% of traffic, reports 0.006% relative lift

Netflix engineers Ying Li, Arjun Rao and Shradha Sehgal tested an LLM-backed ranker using about 40 times fewer Phase 2 labels than the production model, reporting a 0.006% relative lift in an undisclosed core metric.

By · Published

Illustration of Netflix's GenRec LLM-backed ranker reshaping a large streaming catalog, with highlighted tiles and reordered rows showing changed recommendations.

Netflix engineers Ying Li, Arjun Rao and Shradha Sehgal tested an LLM-backed ranker using about 40 times fewer Phase 2 labels than the production model, reporting a 0.006% relative lift in an undisclosed core metric.