Agentic workflows
Agent skills, tools, and MCP integrations that hold up in production — not demos. Built around clear contracts and verifiable steps.
- Agents
- MCP
- Tooling
The through-line
Precision from the military. Pattern recognition from data. Practical systems from AI.
Different chapters, one habit: take something complex and make it move. Here's the arc — and what each part taught me about doing the work.
- The Regiment
75th Ranger Regiment
Where I learned that standards are not negotiable and that a plan only matters if it survives contact. Precision, ownership, and calm under load — the operating habits everything since has been built on.
- Data & NLP
Analytics and early NLP
I followed the data — into energy and venture capital — and got hooked on language models before they were fashionable. The lesson: the hard part is rarely the model, it's framing the right question and trusting the signal.
- Graduate study
MS, Data Science — University of Denver
Formalized the pattern recognition. Statistics, ML, and the discipline to tell a real result from a comfortable one. I came out caring less about novelty and more about systems that hold up.
- AI engineering
Building practical AI systems
Agents, LLM composition, evaluation, and the unglamorous plumbing that makes them dependable. Good AI engineering is mostly judgment about what to automate, what to verify, and what to leave to a human.
- Now
Clinical AI at KaiCare.ai
Remote patient monitoring and care-management systems where the stakes are real. It's the current proving ground for the same principle: turn messy complexity into something a team can actually move on.