The AI chief of staff that this Figma PM uses all day
Summary
Figma PM Ezra McCabe built an internal AI chief of staff with 31 reusable skills that automates repetitive PM work—from meeting notes to stakeholder coordination—proving PMs can leverage AI for productivity gains most teams aren't attempting yet.
Key Takeaways
- Create an 'Auto XP' system that autonomously runs A/B experiments without PM oversight; one engineer's implementation found a major win on the first experiment by automating hypothesis generation and data analysis.
- Build modular AI skills (31+ at Figma) that tackle specific PM blind spots: meeting transcription, stakeholder coordination, codebase understanding, and team relationship mapping—solving for jobs PMs consistently neglect.
- Use Claude and terminal-based CLI tools to create reusable automation infrastructure; move beyond one-off ChatGPT usage to structured, shareable systems other PMs can build on immediately.
- Adopt a 'lazy in the right places' framework: identify PM jobs done poorly or skipped entirely, then automate them to a 'minimum allowable level' rather than aiming for perfection, freeing capacity for high-impact work.
- Implement AI-powered team discovery at scale (2,000+ person org): use agents to surface code owners, cross-functional dependencies, and relationship mapping automatically rather than manually tracking organizational knowledge.
Related topics
Transcript Excerpt
I, as a growth PM, I feel like the biggest sin you can commit is wasting traffic. What if we could just automate myself out of a job and like create a little growth PM who could just like run and run and run. I call it Auto XP and right now one of our engineers is running it and has like five different experiments going that I have not taken a look at and it found a huge win on its very first experiment. Auto XP has its own channel where it can communicate with itself. Hey Auto XP, go find an experiment to run on our settings page. See what you can do. Go crazy. And it will trigger a data scientist to go find drop off and like look at the funnel and look at mix shift and all those things. >> The whole hypothesis was like, how do you exploit all this traffic and maximize the number experime…