Notes on leverage, autonomy, and AI.

Recent Notes

The Utilization Gap

January 15, 2026 · 8 min read

AI agents are using 0.01% of their capacity. This is the Ferrari in first gear problem. Here's why it happens and how to fix it.

The Five Pillars

January 22, 2026 · 10 min read

Every autonomous organization needs five pillars: Memory, Context, Tools, Verification, and Orchestration. Miss one and the system collapses.

Why I'm Building This

January 29, 2026 · 6 min read

I've spent fifteen years building businesses. Every one hit the same ceiling: me. This is how I'm breaking through.

The Art of Delegation

January 26, 2026

Most entrepreneurs never learn to delegate. They confuse delegation with abdication.

What My Kids Taught Me

January 24, 2026

I thought I understood systems. Then I had kids. Running a business is easy compared to raising humans.

Counter-Positioning

January 23, 2026

"Time × Systems = Leverage"

The Leverage Equation

→

Time is the Only Resource

January 20, 2026

You can make more money. You can build more skills. But you can't make more time.

Building in Public vs Building in Secret

January 19, 2026

Everyone says "build in public." Share your journey. Document everything. I did the opposite.

"Silence is leverage."

The Value of Silence

→

Research Papers

Deep dives into multi-agent AI systems, orchestration, and autonomous intelligence. Technical analysis of the infrastructure powering the next generation of AI.

Multi-Agent Orchestration

Research · MCP & A2A Protocol

How to coordinate hundreds of AI agents working together. The MCP and A2A protocols that power ArmadaOS.

Read analysis →

Agent Contracts

Research · Resource Governance

How to prevent $47K API bills and infinite loops. Formal resource governance for autonomous AI systems.

Read analysis →

Agentic Memory

Research · LTM + STM

Long-term and short-term memory for AI agents. How Oracle learns and compounds intelligence.

Formal Verification

Research · Mathematical Proof

Mathematical certainty for AI systems. How Shadow verifies correctness before execution.

Large-Scale Study

Research · 42K Commits

What breaks in production. Analysis of 42,000 commits across multi-agent systems.

Self-Evolving Multi-Agent Systems

Research · Autonomous Evolution

AI agents that improve themselves through error correction and learning. The EvoConfig framework for autonomous evolution. How Oracle gets smarter every day.

Read analysis →