About

I'm Ryan. Since 2010 I've deployed data systems, machine learning, and analytics inside large enterprises. The work has spanned finance, telecom, media, marketing, HR, and credit at organizations including Comcast, TransUnion, DIRECTV, and AT&T. Different industries, different teams, same fundamental challenge: making complex technology produce actual business results in organizations that weren't built for it.

That background shapes everything I do now. I embed directly with enterprise clients to drive AI adoption, moving organizations from isolated experiments to production workflows that compound. The deliverables are working systems: agent architectures, integration patterns, automation pipelines, and deployment playbooks. The job is part engineer, part translator, part operator. The hardest part is rarely the model. It is usually the organizational adaptation around the model.

The domains I've worked across, analytics, business process automation, financial modeling, customer operations, and engineering, give me a lens that most pure-infrastructure people lack. AI adoption is not just a software development story. The same verification gap, the same organizational learning curve, and the same tension between output volume and quality show up everywhere AI enters production work. Code is the most instrumented version. It is not the only one.

My independent research focus is production agentic systems. I'm developing EdgePlane, an open-source control plane for AI agent fleets, built around the concerns that matter in production: persistent identity, lifecycle ownership, inter-agent coordination, and operational observability. Documentation at edgeplane.ai. I also build and operate Aria, a personal AI system that handles research, publishing, and analysis workflows. It is a living testbed for the agent patterns I write about.

PhD trained in Economics and Organizational Psychology from Claremont Graduate University, and a BS in Business Management from CU Boulder. The social science background is not decorative. It shapes how I think about AI adoption, persuasion, decision-making under uncertainty, and why the hard problems in enterprise AI are almost always about people and process, not models.

The writing here is for practitioners doing the actual work. Not tutorials. Field reports, research synthesis, and honest assessments of what works, what does not, and why the gap between AI demos and production systems persists. Follow along on Twitter / X for shorter takes, or subscribe via Substack to get posts by email.