Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Consumer GPU Fleet in Production
This talk covers practical strategies for financing, importing, and operating a consumer GPU fleet in a datacenter, focusing on hardware and infrastructure challenges.
Datacenter GPUs are expensive. Hosting at home is doable, but you deal with power, bandwidth, network availability, and space. How do you scale a GPU homelab in production? Here are all the tactical insights I’ve learned from financing & importing hardware, negotiating contracts, planning, and operating rack inside of Digital Realty for the purpose of hosting AI models.