Compiled this from discussions on r/LocalLLaMA, HN, and various dev forums over the past month. Covers Ollama setup, model selection, hardware requirements, and common pitfalls. Feedback welcome.