§ 04 — AUTHOR
Books I've authored — read them free, download them as PDF or ePub, or grab a copy on Amazon.
Why I write
I've been obsessed with keys, keyboards, and typewriters since I was a kid. My mother used to take me with her to work, where she spent long hours drafting educational materials for schools. I'd sit next to her and watch her fingers move across the keys.
I've been with a keyboard my whole life since. She inspired me, and I chose a path that would let me be as good with the keyboard as she was. These books are what came out of it.
The ultimate guide to building enterprise-level systems from idea to product. The Standard describes how to model, name, test, and structure software so any engineer can walk into any Standard-compliant codebase and feel at home. Free to read on GitHub, forever — also available on Amazon.
The companion to The Standard for the people who build with it: how a team finds its purpose, contributes code, names and reviews its work, keeps its secrets, and measures contribution fairly, whatever the title of the person who made it. People over processes.
Guidelines for C# coding styles: how to name files, variables, methods, and classes, how to lay out declarations, calls, and chains, and when a comment earns its place. Every rule shown as a Do and a Don’t.
A collection of articles from 20 years of experiences in the software industry — on engineering, teams, leadership, and the craft of building things that last.
Learn C# from the ground up — a from-first-principles introduction to the language for engineers starting their journey.
The architecture and economics behind PeerLLM, written up in full.
The vision, architecture, and economics of PeerLLM: a peer-to-peer network where anyone can earn from their idle hardware, their data, and their models, so the gains of AI are shared instead of concentrated in a handful of companies.
How PeerLLM runs inference across community hosts: splitting prompts into dependency graphs, racing hosts under strict deadlines to cut tail latency, and continuing across models when one fails. Scaling intelligence by improving the execution policy, not only the machines or the models.
The data side of PeerLLM. Questions the network cannot answer become knowledge gaps routed to verified experts, whose validated answers are stored with attribution and pay their authors every time they ground a model’s answer.