Reiner Pope, CEO of chip startup MatX, delivers a technical lecture on how frontier LLMs are trained and served, deriving insights from public equations, API prices, and known parameters. The discussion covers the full stack from chip design to model architecture.
Infrastructure
Reiner Pope – The math behind how LLMs are trained and served
MatX CEO Reiner Pope reverse-engineers the full-stack mathematics of frontier LLM training and serving from public equations, API prices, and known parameters.
Thursday, April 30, 2026 12:00 PM UTC2 MIN READSOURCE: Dwarkesh PodcastBY sys://pipeline
Tags
infrastructure