New book is out: The Local AI Handbook It covers all my lessons (and other people's lessons) on building custom GPU rigs at home. All the way from talking to electricians to get the proper wiring and sockets installed, cooling optimization, GPU tradeoffs and the metrics to look at for training vs just quick and easy (and really fast) local inference so you don't have to rely on the cloud anymore. This is hardware deep dive, as well as looking at NVIDIA, Apple, AMD, Intel, etc chips to know whats best for you. I recommend this for anyone looking to know what questions to ask, and to learn about the tradeoffs you likely didn't even know exist in the end to end process. It's my goal to make sure that local ai thrives and that you know you're spending your money correctly when you invest in a rig. I'll be iterating on this one as feedback comes in, so expect frequent updates. This is the first full version I'm happy with making public now. Enjoy :)