A guide to thinking with AI, not just using it.
10 lessons from real projects. No jargon, no hype — just a method that works.
Available on Kindle now (£6.99). Paperback releases 6 August 2026 (£14.99).
A method for working with AI that produces reliable results — not lucky ones.
The difference between AI that produces generic advice and AI that produces something only you could have written is fifteen minutes of setup.
AI is wrong often enough that you need a system for catching it — not a feeling.
The best ideas often come from two AIs disagreeing. Here's how to set that up.
A teacher built a quiz that branched by ability. A paralegal wrote a personal statement that sounded like her. You don't need to know code — you need to know what you want.
Each chapter draws on a project I actually built. These four shaped the book — no team, no budget, no engineering background. AI made them possible.
A publishing platform I built and now run as a product. Upload a Word manuscript, get a print-ready PDF and a Kindle-ready EPUB. AI checks cross-references, citations, glossary terms, and structural fidelity along the way. Different standards packs handle different kinds of book — practical guides, technical references, more to come. Plyego produced this book and the other one.
Tracks portfolios against rules I set when thinking clearly. Flags drift. Records every decision and why. Runs quietly between monthly reviews. The next version will implement investment strategies based on my own research, with proper attribution analysis to see what's actually working — and what's not.
A questionnaire (200+ questions), a locked brief that can't silently drift, a three-layer triage that filters every new listing, and a red-team challenge for every serious candidate. Built to stop me falling for the wrong property. Still running, still doing its job.
A repo I run as my own command centre. It holds my AI stack decisions, my brand strategy, what I'm building and why, and the rules I use to coordinate across two books, a platform, and a day job. AI helps me keep it honest — surfacing inconsistencies, challenging weak rationale, holding me to the cool-off periods I've committed to.
AUTHOR
Author
A quantitative risk modeller (MSc, CFA) who learned to use AI properly without becoming an engineer. He writes for people who keep meaning to start.
Also the author of Crypto Systems Collection Volume 1 (Amazon, 2026).
Ten chapters across three parts. Each chapter ends with a hands-on exercise you can try in any AI tool.
Part 1 · Chapters 1–3
Part 2 · Chapters 4–7
Part 3 · Chapters 8–10
Still reading? You're more curious than most. Start where you like — the book, the exercises, or the opening.