Season 2 · Episode 484
The Silicon Sharing Economy: Inside Serverless GPUs
How do small teams run massive AI models without $50,000 chips? Corn and Herman dive into the hidden plumbing of serverless GPU providers.
My Weird Prompts · Daniel Rosehill
February 4, 202624m 32s
Audio is streamed directly from the publisher (dts.podtrac.com) as published in their RSS feed. Play Podcasts does not host this file. Rights-holders can request removal through the copyright & takedown page.
Show Notes
Ever wonder how a tiny startup can run massive AI models that require hardware costing more than a luxury car? In this episode, Corn and Herman pull back the curtain on serverless GPU providers like Modal and Core Weave to explain the "plumbing" of the modern AI era. They explore the shift from reselling AWS instances to building specialized "Tier Two" data centers, the engineering magic behind sub-second cold starts, and why the "sharing economy for silicon" is the only way for developers to survive the hardware wars of 2026.