About MonkeyRun

I'm Matt. I've spent 25 years building and scaling startup teams — as a developer, architect, CTO, founder, and angel investor. Now, I run a startup studio using AI agents instead of employees.

Why I Built This

AI made coding cheap. That was the easy part. The hard part is what comes next: coordination, structure, and stage gates.

We are living through a fundamental shift in how software is built. But while the cost of writing a line of code has plummeted to near zero, the cost of building a product hasn't.

Why? Because products require teams. They require product management, security audits, marketing distribution, and operational discipline. You still need structure. You still need stage gates along traditional lines to prevent zombie projects.

MonkeyRun is my live laboratory for exploring this new reality. I don't just use AI to write code; I use AI agents to fill specialized roles—founders, PMs, security researchers, and a COO that coordinates them all.

This project is about openly sharing the trials and tribulations of that journey. I document the context overflows, the agent conflicts, and the architectural patterns that actually work when you treat AI as a team rather than a tool.

What You'll Find Here

If you're an early-stage founder who is tight on cash, overwhelmed by your roadmap, or curious about AI but unsure where to start operationally, this site is for you. I share:

• Playbook posts — How to set up multi-agent orchestration from scratch.
• Tool reviews — What actually works when you test the latest AI tools in production.
• Case studies — Real examples from the products I'm building.
• Founder templates — Downloadable, practical, immediately useful files like CONVENTIONS.md and orchestrator setups.
• Failure logs — Honest, specific accounts of when AI agents made things worse.

Why “MonkeyRun”?

The name is the existential question: where does the human fit when AI can do most of the work?

The monkey runs because it has to. Not because the machines can't — but because someone still needs to set the direction, make judgment calls, and decide what's worth building in the first place. The old Russian proverb says it best: “Trust, but verify.”

That's the MonkeyRun thesis — and it's really just another way of saying human-in-the-loop. AI agents can fill most roles on a startup team. But the human in the loop isn't optional — they're the reason the whole thing works. The question isn't whether to use AI. It's how to stay useful when your team doesn't need sleep, doesn't get distracted, and works at the speed of an API call.

What's Next?

I am continuously evolving the HWW-1.5 architecture, my Open Brain memory layer, and my automated PM engines.

If you're a founder drowning in roadmap, exploring multi-agent architectures, or just want to swap notes on building in this new paradigm, my door is open.