Reiner Pope
CEO of Maddox chip startup and former Google TPU architect.
“We've talked publicly about something which we call a splittable systolic array, which is in some sense you can think of as big systolic arrays that can be small systolic arrays as well.”
Source→“We've talked publicly about something which we call a splittable systolic array, which is in some sense you can think of as big systolic arrays that can be small systolic arrays as well.”
Source→“That ratio is actually slightly wrong. It should be like an even bigger, you should get an even bigger speedup than you might otherwise think. Nvidia's product specs have sort of started acknowledging that in B300 and beyond, where the FP4 is three times faster than the FP8. Right. Though it should be 4X.”
Source→“Seven-eighths of the cost is in the reading and writing the register file. And only a tiny fraction of the cost is in the logic unit itself. So this is the problem to solve.”
Source→“The processes that are inside a lot of AI chips actually also have deterministic latency too. Groq has advertised this, TPUs have that in the core as well.”
Source→“CEO of Maddox, former chip architect (likely Google TPU background given depth of TPU knowledge).”
Source→“Today I'm interviewing Reiner Pope, who is CEO of Maddox, which is a new chip startup. Previously, he was doing TPU architecture and many other things at Google.”
Source→“Today I'm interviewing Reiner Pope, who is CEO of Maddox, which is a new chip startup. Previously, he was doing TPU architecture and many other things at Google.”
Source→“From Hopper to Blackwell is mostly just the decision to switch from trays as the form factor... switching to racks as the form factor. That's a product decision.”
Source→“DeepSeek V3 has about 37 billion active parameters and then 700 billion total parameters.”
Source→“DeepSeek mixture of experts has said actually activate more experts but finer grained experts — was a big innovation.”
Source→“Character AI has a blog post talking about that — alternating long and short context — and like in the global context which is really what we're talking about here, global context was shared across all the layers.”
Source→“Jensen math there, but there is at least a genuine 4x increase”
Source→“Reiner Pope, who is CEO of Maddox, which is a new chip startup. Previously, he was doing TPU architecture and many other things at Google.”
Source→AI-extracted from podcast / newsletter / paper summaries. May contain errors.