Start here

Pick your path. We'll take it from there.

Four guided tracks through the library, matched to where you are today. Every path starts from first principles and builds toward working understanding.

Track 00 · The Groundwork

I am starting from scratch

You have not worked with computers, machine learning, or neural networks before, and you would like to begin at the very beginning. No code, no maths, no assumptions about what you already know. Every idea is explained in plain words and simple pictures.

The GroundworkStart from absolute zero. The computer science, machine learning, and neural network ideas AI is built on, explained in plain words with no equations and no numbers, so every AI term makes sense before you meet it.22/22 · Beginner

Then take Track 01 and carry on from there.

Track 01 · The Curious

I'm new to AI

You use ChatGPT or Claude and want to understand what is happening, then get real work out of these tools. No code required.

AI, From ZeroThe ground floor: what a language model actually is, why it works, and why it fails, explained for someone starting from nothing.10/10 · Beginner AI for Business ProfessionalsNo code, no jargon: how to actually use AI at work, the tools, the prompts, the workflows, and the traps.8/8 · Beginner

Then browse the Explainers. Start with how ChatGPT reads a PDF.

Track 02 · The Builder

I build with AI

You write code or ship products and want to use AI tools properly, with prompting that works, retrieval that stays honest, and agents in your terminal.

Prompt Engineering That WorksPast the tips-and-tricks noise: the small set of techniques that measurably change model output, and how to know they worked.7/7 · Intermediate RAG, ProperlyWhy retrieval-augmented generation goes wrong in production, and the specific decisions that keep answers grounded in something real.8/8 · Intermediate Claude Code, MasteredThe agentic coding tool, feature by feature: from first install to workflows that feel like cheating.10/10 · Intermediate MCP in PracticeThe Model Context Protocol, hands on: what it standardizes, how to use servers, and how to build your own.5/5 · Intermediate AI Coding ToolsAgents, copilots, and autocomplete: what each tool actually does, how they think, and how to build a workflow that compounds.6/6 · Intermediate

Then the Playbooks and Protocols articles keep you current.

Track 03 · The Architect

I design AI systems

You are responsible for AI in production, including agents that do not loop forever, retrieval that scales, and evaluation that catches regressions before users do.

AI System DesignHow production AI agents actually work: the reasoning loop, memory, tools, guardrails, evaluation, and the unglamorous engineering that keeps them reliable.14/14 · Advanced Inference InternalsWhat actually happens between hitting enter and seeing tokens: the caches, the batching, and the economics under every AI product.7/7 · Advanced How Vision Models WorkFrom CLIP to diffusion to video: how machines learned to see, draw, and read the visual world.6/6 · Advanced

Then the Deep Dives library goes as far down as you want.

Opening the library

Pick your path. We'll take it from there.

I am starting from scratch

I'm new to AI

I build with AI

I design AI systems

Keep up with AI. One email a week.

Pick your path. We'll take it from there.

I am starting from scratch

I'm new to AI

I build with AI

I design AI systems

Keep up with AI. One email a week.