How We Started a Coop Journey to 10x Me
After 20 years of coding, I realized something: the bottleneck was never skill — it was time. There's only so much one person can build in a day. So I started an experiment: building a team of AI agents to work alongside me, multiplying what I can ship.
This is the story of how the 10x Me journey began.
The Problem
As a co-founder at ELIA, I'm spread thin. Product strategy, coding, business development — everything competes for the same 24 hours. Delegating to humans takes time: hiring, onboarding, managing. And sometimes you just need something done now, not next week.
The answer wasn't more humans. It was smarter agents.
The Setup
I built a small team of AI agents that handle specific tasks: coding, research, documentation, deployment. Each agent has a clear role. Each knows the context. When I'm busy, they keep shipping.
Think of it as a virtual team. I'm the founder and product lead. The agents are the engineers who execute.
What We Learned
- OpenClaw feels magical. From the first prompt, it just works. But the magic has a cost — tokens add up fast.
- Manage tokens from day one. Context is expensive. Be surgical about what you feed the agent. Save memory, summarize history, prune aggressively.
- Context is the hard part. The model remembers, but not how you want it to. Building persistent context across sessions is still something we're figuring out.
- One agent isn't enough. After a few weeks, I realized: I need another agent. Different role, different context. The team needs to grow.
What's Next
This is just the beginning. The goal isn't to replace human work — it's to 10x what one person can accomplish. The team grows. The agents learn. The output compounds.
We'll write more about what we build, what breaks, and what we learn along the way.