Reiner Pope – The math behind how LLMs are trained and served
This podcast episode is a technical deep-dive lecture on the infrastructure and economics of AI inference, focusing on how model architecture, hardware constraints, and batching strategies determine latency, cost, and scalability. It lands by revealing the hidden engineering trade-offs be
Sign up to read the full summary
Free AI-powered recaps of every episode, delivered to your inbox.
Free forever for up to 3 podcasts. No credit card required.
Guest
Reiner Pope
narrative on Dwarkesh Podcast
Mentions
Books, people, and resources mentioned in this episode.
People
Companies
Research & Papers
Topics
Listen to This Episode
More from Dwarkesh Podcast
Why smarter AI models could drive up compute prices 10x
Reiner Pope – Chip design from the bottom up
Dylan Patel – Anthropic & OpenAI will have most of the world’s compute by 2028
Eric Jang – Building AlphaGo from scratch
The next big breakthrough will be AIs learning on the job
Grant Sanderson – AI and the future of math
Related episodes
From other shows on similar topics.
Anjney Midha's Plan to Radically Lower the Price of Compute
Odd Lots
How Companies Are Becoming AI Token Efficient
The AI Daily Brief: Artificial Intelligence News and Analysis
Open Source Wins, AGI Is Here, and Scorsese's AI Toolkit with CEOs of Cerebras & Black Forest Labs
All-In with Chamath, Jason, Sacks & Friedberg
Jeff Dean: The 1% Rule for Building in AI
Y Combinator Startup Podcast
Ep 89: AI Research Legend’s Honest Assessment of Where We Are
Unsupervised Learning with Jacob Effron
Why AI Has No ROI with Paul Kedrosky
Better Offline
Get summaries like this every morning.
Free AI-powered recaps of Dwarkesh Podcast and your other favorite podcasts, delivered to your inbox.
Free forever for up to 3 podcasts. No credit card required.