Y Combinator posts Paper Club video on AI inference, from API requests to GPUs
YC's program lists talks on request lifecycles, open-model tokens, search-based compilers and memory.
TLDR
Y Combinator uploaded an inference-focused Paper Club video on Oct. 24, featuring speakers from Together AI, Ollama, Luminal, SGLang and RadixArk. YC frames the session around a request's path from API to GPU and back, including routing and caching. Dan Fu's opening explains how routing can account for cached work; YC's chapter list identifies the later talks but does not establish their findings.
Combined views
20.1K
5 Sources, first seen ago



