Blog
Notes from the practice.
Essays live here so Google, referrals, and bookings all meet on your domain. The newsletter is the inbox version of the same thinking.
Practice · August 28, 2026
Robotics just leapt a year — from Beijing’s robot games to a $399 duck you can train
The 2026 World Humanoid Robot Games put more than two thousand machines on a track. Microduck puts reinforcement learning on a desk. Together they show where AI in the physical world is heading next.
Read post
Practice · August 4, 2026 · 11 min
GPT-5.6 and Mythos-class Claude: you're probably using them wrong
As of GPT-5.6 and Anthropic’s Mythos 5 era, both models are extraordinary — and most people still trap them in a browser tab. The desktop apps (ChatGPT Work and Codex, Claude Cowork and Code) are where the real work happens.
Practice · July 22, 2026 · 10 min
AI skills: how SKILL.md files make complex work repeatable
A skill is a playbook the model can load on demand. Describe the job, let Claude or ChatGPT draft the SKILL.md, then review it — and watch multi-step work get sharper every time you run it.
Practice · July 8, 2026 · 8 min
Stop scanning. Start reviewing: AI for weekends — and for competitors
A Thursday email of twenty local events, filtered for one family, is the same method a company can use for competitive research: let AI scan, and keep human judgment for the last pass.