ML practitioner slams sudden wave of self-appointed LLM inference gurus on X

trawasthi_ai · x · 2026-09-17

An ML practitioner complains on X that after a few researchers posted about KV Cache, nearly every AI/ML account suddenly turned into an "LLM inference guru." He challenges whether they have ever shipped actual inference optimizations, notes there is a lot to learn before even reaching KV Cache, and argues that positioning yourself as an educator comes with responsibility.

Original post →

More from Fun

Fun channel →