Appears On
Episode Appearances
Dwarkesh Podcast · Apr 29, 2026
Reiner Pope – The math behind how LLMs are trained and served
“Guest delivering technical lecture on LLM training and inference mathematics and hardware constraints”
Roofline analysis for transformer inferenceBatch size optimization and latency-cost tradeoffsMemory bandwidth as binding constraint on inference
View Analysis
Cheeky Pint · Feb 26, 2026
Reiner Pope of MatX on accelerating AI with transformer-optimized chips
“Co-founder and CEO of MatX; former TPU architect at Google; Haskell programmer turned chip designer”
Transformer-optimized chip architectureHBM vs SRAM memory trade-offs in AI inferenceSystolic array design and optimization
View Analysis
TBPN · Feb 24, 2026
Citrini Memo Reactions, Kim K Enters Energy Drinks, Jane Street Sued | Patrick & John Collison, Bill Gurley, James Cadwallader, Scott Wu, Ivan Zhao, Stefano Ermon, Rune Kvist, Reiner Pope, Devansh Pandey
“MadX CEO announced $500M funding for custom AI inference chips”
Citrini memo reactions and AI economic impactStripe Series C funding and agentic commerce strategyAI agent enterprise adoption and deployment
View Analysis