Reiner Pope – Chip design from the bottom up
AI chips are optimized to maximize compute efficiency by minimizing data movement, with systolic arrays enabling massive matrix multiplication efficiency through spatial computation and local storage of weights. The clock cycle synchronizes all chip operations, and design trade-offs center
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
interview on Dwarkesh Podcast
Mentions
Books, people, and resources mentioned in this episode.
People
Companies
Products & Tools
Research & Papers
Topics
Listen to This Episode
More from Dwarkesh Podcast
Reiner Pope – The math behind how LLMs are trained and served
Why smarter AI models could drive up compute prices 10x
Dylan Patel – Anthropic & OpenAI will have most of the world’s compute by 2028
The data black hole at the center of AI
The next big breakthrough will be AIs learning on the job
Eric Jang – Building AlphaGo from scratch
Related episodes
From other shows on similar topics.
How CoreWeave Sees the Market for Compute Right Now
Odd Lots
Jeff Dean: The 1% Rule for Building in AI
Y Combinator Startup Podcast
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
How Companies Are Becoming AI Token Efficient
The AI Daily Brief: Artificial Intelligence News and Analysis
Ep 89: AI Research Legend’s Honest Assessment of Where We Are
Unsupervised Learning with Jacob Effron
The history and future of AI at Google, with Sundar Pichai
Cheeky Pint
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.