Subvolts take Understand how AI hardware works, from logic gates to GPUs, with a former Google TPU architect. May 22, 2026 - Developers
- Learn foundational hardware principles that underpin AI development and optimization efforts.
- Business
- Understand the hardware constraints and innovations driving AI capabilities and costs.
A solid interview with a former Google TPU architect explaining fundamental chip design principles relevant to AI hardware.
Reiner Pope – Chip design from the bottom up
Audio streams from Dwarkesh Patel. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Dwarkesh Patel interviews Reiner Pope on the Dwarkesh Podcast about chip design. Pope, CEO of MatX and former Google TPU architect, explains how chips work from basic logic gates to complex architectures like GPUs and TPUs. The episode explores the foundational principles behind modern computing hardware.
This episode breaks down how computer chips are built, starting with simple on/off switches. It explains how these build up to create powerful processors like those used in AI, and even compares them to the human brain. It's for anyone curious about the physical foundations of modern technology.
This episode offers a deep dive into chip architecture for those interested in the hardware powering AI. Listeners will learn how basic logic gates build up to complex processors like GPUs and TPUs. It's for engineers, hardware enthusiasts, and anyone curious about the physical infrastructure behind AI advancements. Understanding chip design is crucial as frontier labs like Google Gemini and OpenAI push the boundaries of AI, requiring ever more specialized and efficient hardware.
- HostDwarkesh Patel
- GuestReiner PopeCEO, MatX
- Guest: Reiner Pope
- Title: Chip design from the bottom up
- Show: Dwarkesh Podcast
- Length: 81 minutes
- Guest's Previous Role: Worked on software efficiency, compilers, and TPU architecture at Google
- Guest's Current Role: CEO of MatX
New blackboard lecture with Reiner Pope: how do chips actually work - starting with basic logic gates, and working up to why GPUs, TPUs, FPGAs, and the human brain each look the way they do.
Reiner is CEO of MatX, a new chip startup (full disclosure - I’m an angel investor). He was previously at Google, where he worked on software efficiency, compilers, and TPU architecture.
Watch this one on YouTube so you can see the chalkboard. Read the transcript.
Sponsors
* Crusoe was one of only five GPU clouds that made the gold tier in SemiAnalysis' most recent ClusterMAX report. Gold-tier providers like Crusoe delivered 5-15% lower TCO than silver-tier clouds, even with identical GPU pricing. This is because optimizations like early fault detection and rapid node replacement don't necessarily show up in the sticker price, but still matter a ton in the real world. Learn more at crusoe.ai/dwarkesh
* Cursor is where I do most of my work—from reading research papers to visualizing technical concepts to coding up internal tools for the podcast. Most recently, I used it to build two different review interfaces for my essay contest, one that anonymizes submissions for scoring and another that lets me see applicants' essays next to their resumes and websites. Whatever you're working on, you should try doing it in Cursor. Get started at cursor.com/dwarkesh
* Jane Street let me ask Ron Minsky and Dan Pontecorvo, two senior Jane Streeters, a bunch of questions about how they use AI. We discussed everything from the types of models they're training to how they think about the future of trading to why they're more bullish than ever on hiring technical talent. You can watch the full conversation and learn more about their open positions at janestreet.com/dwarkesh
Timestamps
00:00:00 – Building a multiply-accumulate from logic gates
00:16:31 – Muxes and the cost of data movement
00:26:10 – How systolic arrays work
00:39:11 – Clock cycles and pipeline registers
00:51:51 – FPGAs vs ASICs
01:03:25 – Cache vs scratchpad
01:07:27 – Why CPU cores are much bigger than GPU cores
01:12:00 – Brains vs chips
01:15:33 – A GPU is just a bunch of tiny TPUs
This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.dwarkesh.com
- logic gates
- Reiner Pope
- Basic electronic circuits that perform simple logical operations, like 'and' or 'or'.
- GPU
- Reiner Pope
- Graphics Processing Unit, a specialized processor good at handling many tasks at once, often used for AI.
- TPU
- Reiner Pope
- Tensor Processing Unit, a specialized processor designed by Google for machine learning tasks.
- FPGA
- Reiner Pope
- Field-Programmable Gate Array, a chip that can be reconfigured after manufacturing for specific tasks.
1 h 371 h 201 h 37AI researchers debate how close we are to recursive self-improvement
WithJohn Schulman, Beren Millidge, Charlie O’Neill
2 h 21Same guests, other shows
2 h 14
56 min
Questions people ask
- What is this episode about?
- This episode is a blackboard lecture explaining chip design from basic logic gates up to GPUs, TPUs, and FPGAs, featuring Reiner Pope.
- Who is Reiner Pope?
- Reiner Pope is the CEO of MatX and previously worked on software efficiency, compilers, and TPU architecture at Google.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Dwarkesh Patel.