38:46
Greg Isenberg
I'm Obsessed With Local AI. Here's Why
A 39 minute solo case for running open models on hardware you own, rebuilt in full. Isenberg maps local AI into four pieces (the model, Hugging Face as the warehouse, LM Studio and Ollama as the software, the workflow you wrap around it), decodes the vocabulary that keeps non technical founders out (parameters, tokens, context window, quantization, GGUF), then walks the whole Gemma 4 lineup with real sizes and memory figures, three install paths, and a hardware cheat sheet by RAM tier. The back half is the business case: a first workflow on a folder of support tickets, the cheapest useful eval, and three startup ideas specified down to the Customer, the flags, and the go to market.