Why this site exists

I’ve spent the last year and a half running the delivery system for a remote product team: release trains, sprint cadence, capacity planning, and the reporting that makes risk visible before it becomes a slip.

Somewhere in there, the job changed. A lot of the work I used to do by hand, like writing changelogs, building release checklists, and reconstructing where tickets actually sat in QA, I now do by building small tools with Claude Code and letting them run. The interesting part isn’t that the AI can do it. It’s figuring out which parts to hand over, how to verify the output, and how to make the whole thing durable enough that the team can pick it up without me.

That’s what I want to write about here. Expect posts on:

• Process design for small remote teams
• What AI-assisted operations actually look like day to day
• The product management side of running launches end to end

The Projects section is where I’ll keep the things I’ve built. Some are polished, some are experiments. All of them are real.