Articles on autonomous PM workflows, context engineering,
and Claude Code for product teams.
MCPs look great in demos. But at team scale, they cost 6x more in token overhead and add failure points. Here's why production PM teams switch to API skills.
How AI agent teams process user interviews in parallel — turning 10 transcripts into structured opportunity trees, theme maps, and evidence tables in 15 minutes instead of 10 hours.
Every Claude Code setup guide is written for one PM. This is the guide for rolling it out to a product team — shared context, consistent skills, and onboarding that takes 30 minutes instead of 3 weeks.
Anthropic launched scheduled tasks and /loop in Claude Code in March 2026. This turns PM operating systems from toolkits into runtimes. Here's what autonomous PM workflows look like in practice — and how to set up your first one.
The best engineering CLAUDE.md files include workflow orchestration, self-improvement loops, and verification gates. Here's what that same rigor looks like for product management.
I used ChatGPT for everything — PRDs, competitive analysis, interview synthesis. Then I switched to Claude Code. The difference isn't the model. It's persistent context, file-based output, and agent teams that changed how I do PM work.