FeaturedLive · 19,000+ servers · public drift ledger
I built mcpindex.ai, the trust-to-act layer for agent tool use.
An agent is about to call a third-party MCP tool. mcpindex answers one question: is this safe to call right now? 19,000+ servers indexed, 18,500+ judged against a published methodology, safety-relevant contract changes recorded in a public drift ledger, and a gate your agent can install from npm or PyPI in one line. The case study covers what shipped, what is deliberately not claimed yet, and why the trust-verdict seat has to be third-party.
Read the case study →
I BUILD (THE OPERATOR)I'm a product leader who still ships. I architect agentic AI systems at work, and after hours I run mcpindex.ai, publish measurement datasets on the MCP ecosystem, and tune the local inference cluster that serves my own private AI tools. Technical fluency is not a garnish; it is the filter that separates what demos well from what survives production.
I PRIORITIZE (THE ALLOCATOR)Strategy is the art of saying 'no' to the good so you can say 'yes' to the great. I look at time, talent, and technology as capital. Whether analyzing the ROI of an AI feature or exploring the biological logic of the human brain with my brother, I am focused on maximizing impact.
I bridge the gap between "Can we build it?" and "Should we build it?" I'm here to create systems that don't just work, but scale—driven by logic and built with intent.