I design self-improving, self-governing multi-agent systems — with the guardrails, observability, and cost discipline that make autonomy something you can actually trust. Not prototypes. Production.
The same engineering discipline, framed for what you care about.
Multi-agent systems that collect data, analyze it, propose the next experiment, and keep the loop safe and supervised. Mapped directly to a self-driving lab.
A productized build funnel — intake, scoping, blueprint, and delivery — that turns an idea into a shipped, governed application without the agency overhead.
Installable PWAs shipped end-to-end — from a daily intelligence brief to an AI story-and-image universe. Real products, live in production.
Whether it's a national-lab pitch, a client's SaaS, or a consumer app, the same backbone shows up: design it, govern it, ship it, repeat it. The work below all runs on it.
A production agent platform: ~30 engines, 5 concurrent learning loops, a curated LLM gateway, a self-healing escalation bridge, and a nightly "Dream Cycle" that proposes its own improvements. It's designed so a human works about 10 hours a week while the agents carry the load.
Full breakdown ›Every unit of work passes a contract, an expert review panel, and a hard gate before it touches production — not by reminder, by design. Behavioral discipline, ROI economics, and kill-switch rollback are built in.
How I gate production AI ›Autonomous systems, a custom application, or just comparing notes — I'd welcome the conversation.