[{"@context":"https:\/\/schema.org\/","@type":"BlogPosting","@id":"https:\/\/blog.terabox.com\/how-ai-chips-work-internal-architecture#BlogPosting","mainEntityOfPage":"https:\/\/blog.terabox.com\/how-ai-chips-work-internal-architecture","headline":"Inside the Silicon: How AI Chips Actually Work","name":"Inside the Silicon: How AI Chips Actually Work","description":"\ud83d\udcfa Today’s recommended deep-dive video: https:\/\/www.youtube.com\/watch?v=oIk3R-sMX5o Inside the Silicon: How AI Chips Compute at ScaleThe Atomic Unit of AI: The Multiply-AccumulateThe Hidden Tax: Data Movement and MuxesSystolic Arrays: Baking Loops into MetalTiming, Throughput, and DeterminismKey TakeawaysQ&A Inside the Silicon: How... ","datePublished":"2026-06-18","dateModified":"2026-06-18","author":{"@type":"Person","@id":"https:\/\/blog.terabox.com\/author\/flextech-admin\/#Person","name":"flextech-admin","url":"https:\/\/blog.terabox.com\/author\/flextech-admin\/","image":{"@type":"ImageObject","@id":"https:\/\/secure.gravatar.com\/avatar\/ad516503a11cd5ca435acc9bb6523536?s=150&d=mm&r=gforcedefault=1","url":"https:\/\/secure.gravatar.com\/avatar\/ad516503a11cd5ca435acc9bb6523536?s=150&d=mm&r=gforcedefault=1","height":96,"width":96}},"publisher":{"@type":"Organization","name":"terabox","logo":{"@type":"ImageObject","@id":"http:\/\/blog.terabox.com\/wp-content\/uploads\/2021\/11\/logo\u4ea7\u54c1\u540d-\u7ad6\u7248.png","url":"http:\/\/blog.terabox.com\/wp-content\/uploads\/2021\/11\/logo\u4ea7\u54c1\u540d-\u7ad6\u7248.png","width":900,"height":900}},"image":{"@type":"ImageObject","@id":"https:\/\/blog.terabox.com\/wp-content\/uploads\/2026\/06\/maxresdefault-4.jpg","url":"https:\/\/blog.terabox.com\/wp-content\/uploads\/2026\/06\/maxresdefault-4.jpg","height":720,"width":1280},"url":"https:\/\/blog.terabox.com\/how-ai-chips-work-internal-architecture","video":{"@context":"http:\/\/schema.org\/","@type":"VideoObject","@id":"https:\/\/www.youtube.com\/watch?v=oIk3R-sMX5o#VideoObject","contentUrl":"https:\/\/www.youtube.com\/watch?v=oIk3R-sMX5o","name":"Chip design from the bottom up \u2013 Reiner Pope","description":"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.\n\nReiner is CEO of MatX, a new chip startup (full disclosure - I\u2019m an angel investor). He was previously at Google, where he worked on software efficiency, compilers, and TPU architecture.\n\n\ud835\udc04\ud835\udc0f\ud835\udc08\ud835\udc12\ud835\udc0e\ud835\udc03\ud835\udc04 \ud835\udc0b\ud835\udc08\ud835\udc0d\ud835\udc0a\ud835\udc12\n* Transcript: https:\/\/www.dwarkesh.com\/p\/reiner-pope-2\n\n\ud835\udc12\ud835\udc0f\ud835\udc0e\ud835\udc0d\ud835\udc12\ud835\udc0e\ud835\udc11\ud835\udc12\n* 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 https:\/\/crusoe.ai\/dwarkesh\n\n* Cursor is where I do most of my work\u2014from 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 https:\/\/cursor.com\/dwarkesh\n\n* 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 https:\/\/janestreet.com\/dwarkesh\n\n\ud835\udc13\ud835\udc08\ud835\udc0c\ud835\udc04\ud835\udc12\ud835\udc13\ud835\udc00\ud835\udc0c\ud835\udc0f\ud835\udc12\n00:00:00 \u2013 Building a multiply-accumulate from logic gates\n00:16:20 \u2013 Muxes and the cost of data movement\n00:25:59 \u2013 How systolic arrays work\n00:39:00 \u2013 Clock cycles and pipeline registers\n00:51:40 \u2013 FPGAs vs ASICs\n01:03:14 \u2013 Cache vs scratchpad\n01:07:16 \u2013 Why CPU cores are much bigger than GPU cores\n01:11:49 \u2013 Brains vs chips\n01:15:22 \u2013 A GPU is just a bunch of tiny TPUs","thumbnailUrl":["https:\/\/i.ytimg.com\/vi\/oIk3R-sMX5o\/default.jpg","https:\/\/i.ytimg.com\/vi\/oIk3R-sMX5o\/mqdefault.jpg","https:\/\/i.ytimg.com\/vi\/oIk3R-sMX5o\/hqdefault.jpg","https:\/\/i.ytimg.com\/vi\/oIk3R-sMX5o\/sddefault.jpg","https:\/\/i.ytimg.com\/vi\/oIk3R-sMX5o\/maxresdefault.jpg"],"uploadDate":"2026-05-22T16:11:28+00:00","duration":"PT1H20M20S","embedUrl":"https:\/\/www.youtube.com\/embed\/oIk3R-sMX5o","publisher":{"@type":"Organization","@id":"https:\/\/www.youtube.com\/channel\/UCXl4i9dYBrFOabk0xGmbkRA#Organization","url":"https:\/\/www.youtube.com\/channel\/UCXl4i9dYBrFOabk0xGmbkRA","name":"Dwarkesh Patel","description":"Deeply researched interviews\n","logo":{"url":"https:\/\/yt3.ggpht.com\/lG-z7sTfhFIW2Ne1oXMHvXMXyZSaA02_I17gUel0GAEj7OypsSHQ7PE91Vp4bTbpm3PTIAWJdko=s800-c-k-c0x00ffffff-no-rj","width":800,"height":800,"@type":"ImageObject","@id":"https:\/\/www.youtube.com\/watch?v=oIk3R-sMX5o#VideoObject_publisher_logo_ImageObject"}},"potentialAction":{"@type":"SeekToAction","@id":"https:\/\/www.youtube.com\/watch?v=oIk3R-sMX5o#VideoObject_potentialAction","target":"https:\/\/www.youtube.com\/watch?v=oIk3R-sMX5o&t={seek_to_second_number}","startOffset-input":"required name=seek_to_second_number"},"interactionStatistic":[[{"@type":"InteractionCounter","@id":"https:\/\/www.youtube.com\/watch?v=oIk3R-sMX5o#VideoObject_interactionStatistic_WatchAction","interactionType":{"@type":"WatchAction"},"userInteractionCount":171001}],{"@type":"InteractionCounter","@id":"https:\/\/www.youtube.com\/watch?v=oIk3R-sMX5o#VideoObject_interactionStatistic_LikeAction","interactionType":{"@type":"LikeAction"},"userInteractionCount":6666}]},"about":["Insights","\u300eEnglish\u300f","Instructions"],"wordCount":1771,"keywords":["sync"]},{"@context":"https:\/\/schema.org\/","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Inside the Silicon: How AI Chips Actually Work","item":"https:\/\/blog.terabox.com\/how-ai-chips-work-internal-architecture#breadcrumbitem"}]}]