English
The Fluency Map is a mechanistic glossary covering roughly 80 concepts across frontier and agentic AI — built around a single rule: never just define a term, explain why it exists and what it causes. Rather than a flat A–Z reference, entries are organized into six causal strands that build on each other: Compute & Hardware, Model Architecture & Representation, Training, Inference & Serving, Agentic AI, and Economics & Industry — tracing a line from physical constraints (GPUs, chips) up through how those constraints shape model design, training, real-world deployment, and the business narratives that follow. A closing synthesis chapter demonstrates the payoff of this structure by unpacking a single compressed sentence ("reasoning models shift compute from training toward inference") using concepts from every strand at once, showing how the pieces interlock into fluent, owned understanding rather than memorized definitions. The book draws on Kurt Fischer's Dynamic Skill Theory as its organizing philosophy — the goal is concepts that become "sticky," automatically recognized without translation, rather than borrowed vocabulary repeated without comprehension. Intended as a study tool f
Okumak, EPUB indirmek veya dinlemek için ücretsiz hesabınla giriş yap.
artificial intelligence · machine learning · large language models · LLM · frontier AI · agentic AI · AI agents · AI infrastructure · GPU computing · neural network architecture