Volume 1 — The Informational/Epistemic Layer
How models think, how meaning is steered, and how state becomes usable.
Reports
AI-ENG-A — Model Steering: Harness Engineering, Prompt Semantics & Adaptation Choice
Distills four years of bleeding-edge prompting practice into systematic model steering. Covers prompts as probabilistic behavioral interfaces, performance seeds, semantic traction per token, operating stance, demonstrations, runtime adapters, harness design, instruction hierarchy, task framing, output contracts, and the strategic choice between prompting, RAG, memory, fine-tuning, distillation, and tool use.
AI-ENG-B — Context Architecture: State Management & The Tenure Principle
Treats context as a managed state layer rather than a bag of semantically similar text. Covers session and persistent state, scope isolation, authority, temporal validity, contradiction, fact-to-instruction conversion, context compilation, progressive procedural-capability loading, and selected/omitted assembly traces.
AI-ENG-C — The Economic Physics of Inference: Tradeoffs, Cost Attribution & System Margins
Defines the physical and financial laws of AI systems. Covers the tradeoffs among latency, quality, reliability, cost, throughput, context size, model class, and infrastructure. Shifts cost thinking from “model price” to granular attribution by feature, workflow, tenant, user journey, and business outcome.