Education & Learning

Learn Without the Freakout

AI is not magic, not a threat, and not as complicated as the headlines suggest. These resources are designed to make it genuinely understandable — for builders, operators, and decision-makers.

Featured Resource

AI Without Freakout

Horizons & Chunks — a practical framework for understanding AI

A structured guide to understanding AI systems without the hype, fear, or oversimplification. Covers how to think about AI capabilities in terms of horizons (what's possible now vs. later) and chunks (how to break down AI problems into manageable pieces). Designed for people who need to make real decisions about AI — not just read about it.

AI LiteracyFrameworksDecision-MakingPractical AI

Learning tracks

Structured paths through the material

01

Token Economics

Understanding how LLMs consume tokens, why it matters for cost, and how compression changes the equation.

What is a token?
5 min read
Why token count drives your AI bill
8 min read
Compression without quality loss — how QCP works
12 min read
Rate-distortion theory for practitioners
15 min read
TokensCostCompressionQCP
02

AI Architecture Patterns

How to structure AI systems that are safe, auditable, and cost-efficient at scale.

The Step Zero principle — gate before you act
10 min read
Inference routing — when to use which model
12 min read
Self-correcting pipelines — judge, drift, correct
15 min read
Audit chains and idempotency in agentic systems
10 min read
ArchitectureSafetyRoutingAgentic AI
03

AI for Business Operators

Practical frameworks for evaluating, deploying, and measuring AI in real business contexts.

AI receptionist vs. AI agent — when does the upgrade pay?
8 min read
How to read an AI cost model
10 min read
What to ask before buying any AI tool
6 min read
Measuring AI ROI without lying to yourself
12 min read
BusinessROIEvaluationDeployment

Reference materials

Downloads & quick references

Reference Sheet

Token Architect One-Pager

A single-page reference for the Token Architect framework — how to think about token budgets, compression strategies, and cost optimization in one place.

Token ArchitectureReferenceQuick Guide
Download PDF
Technical Paper

QCP Theoretical Foundation v2

The formal theoretical grounding for QCP — rate-distortion theory, hard invariants vs. soft fidelity, and the Calabi-Yau analogy explained. Tightened for technical scrutiny.

QCPTheoryRate-DistortionFormal Methods
Read Paper
Data Sheet

Marcie AI Cost Model

Full cost comparison dataset: AI receptionist vs. agent across 50–1,000 calls/month and 0–10 workflow integrations. Includes break-even calculations per tier.

Cost DataMarcie AIBreak-EvenSpreadsheet
View Analysis

Want a guided walkthrough?

We run workshops and consulting engagements for teams that want to go deeper. No jargon, no hype — just the actual work.

Book a Session