6-1 What Is an LLM A Working Engineers Mental-Model Lecture

If you've been with us through Phase 1, you've built a serious AI-assisted workflow — you've generated code, refactored legacy systems, and shipped pipelines. Phase 2 is different. We are going to stop using AI as a tool that lives somewhere else and start building AI that lives on your own machine. Module 6 is the foundation of that — local LLMs and RAG architectures. And this first lecture is the foundation of the foundation. We are going to define exactly what a large language model is in working-engineer terms. Not philosopher terms, not researcher terms — engineer terms. Because the moment you understand the moving parts, every later decision you make in this module gets easier. So let us strip the magic away.