2 ms·
This is a JVM infrastructure optimization, not a recommendation architecture breakthrough. At Netflix scale, most of the field runs GPU-accelerated ANN search (
by 7777777phil 8mo ago
This is a JVM infrastructure optimization, not a recommendation architecture breakthrough. At Netflix scale, most of the field runs GPU-accelerated ANN search (FAISS, ScaNN) for candidate retrieval.
- oofbey 8mo agoFirst mover disadvantage: you need to invent a way of doing things (ANN) before any decent tools have been built. 10 years later the world is laughing at you for using Java to do vector math.
- lern_too_spel 8mo agoFrom the sounds of it, they're looking for items that are different from their watch history among some candidate set, which vector dbs aren't optimized for.