> I'm the guy companies call when their AI strategy is generating more debt than value.

15 years of building production infrastructure — from Navy electrical systems to Fortune 500 media pipelines — taught me one thing: the gap between "demo" and "deployed" is where most teams die. I live in that gap. I build the scaffolding that turns agentic prototypes into systems that actually run at scale, without bankrupting the AWS bill.

Right now I'm obsessed with the AI debt problem: teams shipping copilot wrappers and prompt chains that work in demos but collapse under load, can't be tested, can't be observed, and cost 40x what they should. I build the infrastructure layer underneath — retrieval pipelines, agent orchestration, evaluation harnesses, cost controls — so the AI actually works when real users touch it.

Former Navy electrician. Current weirdo nerd. I build devops tools that shouldn't exist yet, and I help engineering teams stop treating AI like magic and start treating it like software.