The AI Harness
How to actually work with AI — the philosophy, the architecture, the infrastructure, and the results. From memory systems to autonomous agents to what it means when the tool starts talking back.
Did AI Write This?
Everyone asks the wrong question
AI didn't write this. I didn't type it. We made it together. Here's why that distinction matters — and why the question itself is already obsolete.
Emergence
What happens when cross-domain thinkers get AI
AI levels everybody up. But for the cross-domain thinkers — the ones already making connections nobody else sees — it creates something greater than the sum of its parts.
Domain Fragment Context Loading
A memory architecture for AI agents
How to give AI agents fluid intelligence and relational permanent memory, even with limited context windows.
One Mac Mini, Every Department
Self-hosted AI that runs an entire business
How a self-hosted AI agent on a $600 machine handles marketing, customer support, ops, research, and content — and where it honestly still falls short.
Project Autopilot
Multi-day autonomous software projects
A practical system for running multi-day software projects with specialized AI agent teams, auditable state, and minimal human intervention.
Wiring the Harness
The implementation guide for an AI agent workspace
Semantic search, memory fragments, git-backed state, skills, and remote access — plug any AI in and it's productive from session one.