“One person who knows AI will outperform an entire startup team.”
I’m the person. The team was Google.
I ran AI agents across 16 public repos and merged 3,913 PRs in 30 days.
Google AX’s visible team, same window: 52 total.
75.25x.
Most AI benchmarks test one model on one task at a time.
Wrong unit.
The real test is whether one person can keep 16 repos shipping with agents—from code and review to releases, recovery, and publishing.
30 days straight. Every PR public.
The model alone isn’t the advantage.
The system that runs the agents is.
The one-person AI company is already here — in China.
https://t.co/N8daGAldn9
Most AI benchmarks test one model on one task.
Wrong unit.
Can one person keep 16 repos shipping with agents—across code, review, release, recovery, and publishing?
3,913 merged PRs. 30 days. All public.
The model isn't the advantage.
The system running the agents is.
“One person who knows AI will outperform an entire startup team.”
I'm the person. The team was Google.
AI agents: 3,913 merged PRs in 30 days.
Google AX’s visible team, same window: 52.
75.25x.
The one-person AI company is already here — in China.
https://t.co/gkc8T2KNdC
I use several coding agents across long-running work.
Some finish. Some stop halfway. Some only look finished until I check.
I built Kungfu so the work—not the chat—survives sessions and handoffs.
Open-source alpha: https://t.co/EQLvaRJJcN