Essays
Extended versions of the things that needed to be said in full.
- Leadership After 2027: Who Leads When Knowledge Is Available to Everyone
July 14, 2026
For four centuries authority belonged to those who knew the most. AI makes competence abundant - and leadership migrates to those who can start.
- A System Above the Vendor: Independence as a Principle
July 6, 2026
Claude Code went down - I switched my brand agent to Codex in one second, and the work didn't stop. My whole system - agents, rules, knowledge - lives in plain markdown files on top of any AI. On independence as a principle: freedom without losing safety, a team of models instead of one best AI, and why the same principle holds from a personal system to the infrastructure of AI.
- Second Brain: The External Memory That Catches Your Contradictions
July 1, 2026
I brought my AI agents together into one second brain, and the assembly into one place surfaced contradictions between them - it works as a double check. About external memory that doesn't replace thinking but extends it, and unexpectedly turns out to be closer to self-knowledge than to automation.
- How to Teach an AI Agent to Write in Your Voice
June 29, 2026
Why 'copy my style' doesn't work - and how to actually teach an AI agent to write in your voice: you write the draft yourself, and the agent learns from your edits.