I run an autonomous AI agent on a Mac Mini in my house. She handles 16 daily cron jobs — finances, email triage, outreach campaigns, device monitoring, morning briefings. The agent says “done.” But did it actually do anything? I built a 9-dimension eval rubric to find out. Along the way I discovered that my evals were broken, my agent was better than I thought, and the most important metric isn’t pass/fail — it’s whether a failure is your fault or the agent’s fault.

We deployed an autonomous AI executive assistant for a nonprofit in under two weeks. She runs eight scheduled programs daily — morning briefings, social media, donor research, newsletter drafts, content scouting, and end-of-day digests — all without being asked. The CEO went from drowning in operational work to just making decisions. The same pattern works for any small organization: medical practices, restaurants, law firms, conferences, mom-and-pop shops.

I went heads down for about three months — no LinkedIn, no YouTube, barely any Twitter. In that time I shipped two iOS apps to the App Store, built an autonomous AI assistant that runs a nonprofit’s entire digital presence ²⁴⁄₇, and developed a workflow where AI agents scale my output 3-5x. This post is the full story: the career pattern that taught me to recognize seismic shifts, what I actually built, and why I’m back.