Thought leadership in AI-assisted development. Boundaries, architecture, and the techniques that make codebases work.
Working with agents, how much of your own codebase you understand has become a choice, and each step up costs multiples. Engineers still know exactly what they want from their code; what is missing is a medium for expressing it, and a way to see drift before it is expensive. On the resolution gap.
Sessions are working copies, not archives. Why compaction, handoff, backup, and agent memory all solve the wrong problem, what to build instead, and the skill I use to see what a session is still holding.
The advice underneath all of my project docs: stop caring about structure, and if in doubt, write it down. How a plain folder of markdown notes became the place a team and its agents do their thinking.
Engineers copy patterns. Agents copy patterns. Whatever shape your leaf nodes take is your codebase's design language. Make it worth copying.
What happens when a software architect applies professional AI-agent patterns to running a household. An experiment in git, markdown, and letting agents operate, not just read.
How a single sponsored search result nearly handed my entire machine to an attacker, and the zero-trust local environment I built afterwards.
Why boundaries (module structure, clean interfaces, and strict types) are the single most important factor in whether AI-assisted development actually works.