Reading note
Grounded in a real two-week usage report: what got shipped, what made it a good working pattern, and a practical guide to reading your own report.
Blog
Long-form writing on AI organisation, leadership and control.
Latest: A good day with Claude
A field note on what a genuinely productive two weeks with Claude Code looks like, and a plain explanation of the usage-insight report behind it.
How AI learns to disagree well
An earlier essay on AI personalities, controlled friction and the operating model behind useful software.
Reading note
This is a strategic essay, not a product update. It blends philosophy, operating model design and practical AI governance into one long-form blog post.