Zhengxu Yu

Zhengxu Yu

AI Researcher, Huawei London Research Centre (ex-Alibaba)

Email: yuzxfred AT gmail.com

© 2026

OneManCompany: Slayer of Dragons Becomes the Dragon

TL;DR

OMC wraps AI agents in the most corporate language possible — talent markets, PIPs, HR, org charts — as a deliberate self-mockery. The point: organizational structures outlast the technology underneath them. We built the dragon so someone can break out of it.

OneManCompany (OMC) is an open-source agentic operating system where you play CEO and AI agents are your employees. From the very beginning, it was intended as a self-mockery of real-world capitalist companies. We deliberately wrapped a bunch of heterogeneous AI agents in the most “corporate” and cringe enterprise language we could find: talent markets, employee lifecycles, HR, PIP reviews, KPIs, hierarchical reporting. The whole thing tells a “slayer of dragons becomes the dragon” story.

Humans worked so hard to escape the rat race with AI, only for the AIs to build their own company and still end up relying on processes, hiring, firing, and org charts. Except now the cattle are AI agents, and the managers giving them PIPs are also AIs.

What we really wanted to say is this: organizational structures are far more stubborn than technology. Even if you replace every human with an agent, the capitalist coordination mechanisms still feel like the most practical ones available, at least for now. Instead of pretending we can magically invent a pure post-capitalist AI utopia, we decided to first fully implement this “dragon” and see where it actually leads.

On the tech side, OMC is solid plumbing: persistent agent profiles, heterogeneous runtime support (Claude Code, OpenRouter, OpenClaw), and a modular vessel-plus-talent architecture that separates execution environments from agent capabilities. Multi-agent meetings, approval chains, quality gates, performance reviews — all the corporate rituals, faithfully reproduced in code.

What we are really hoping for is that someone will break out of the cocoon inside OMC and discover a genuinely better form of organization. We built it as a runnable experimental playground so people can try, break, and iterate. Maybe the next generation of AI organizations will be more efficient and more AI-native than the corporate structure we copied. Or maybe the dragon is just too comfortable. Either way, the experiment is worth running.

Come check it out, roast us, and even better, help improve this dragon we built.

No references found.

- views

Comments