I'm a Senior Cybersecurity Systems Engineer with over 25 years of experience in the industry and government, now building production agentic AI systems with defense-grade rigor.
My career spans aircraft avionics, acquisitions, systems engineering, test & evaluation, and weapon system cybersecurity — through assignments with the NSA, ACC, AFSOC, AFTC, AFLCMC, PACAF, and Percival Engineering. I graduated from the NSA's Computer Network Operations Development Program (CNODP) and have led cybersecurity operations for up to 1,800 Airmen across multiple NSA locations.
That foundation — reverse engineering cyber-physical systems, conducting vulnerability assessments, and operating in environments where failure is not an option — shaped how I think about AI. Every autonomous decision needs governance. Every workflow needs structured oversight. Every system needs to degrade gracefully when the model gets it wrong.
I've built multi-agent platforms from scratch: agent harnesses with structured communication architectures, intelligent LLM routing layers that minimize cost while maximizing capability, governance dashboards that monitor agent decisions in real time, and autonomous research agents that operate on long time horizons with minimal human intervention. I also research local model fine-tuning — building pipelines that take agent execution data, curate it into training sets, and fine-tune open-weight models on on-premise GPUs so that routine tasks graduate from cloud inference to local execution at a fraction of the cost.
I care deeply about making AI systems that are reliable, interpretable, and steerable — not just capable.
Starting May 2026, I'll be pursuing a PhD in AI/ML, focused on advancing the science of agentic AI systems.
The best AI systems are engineered with the same rigor as the physical systems they analyze. I approach agent design the way a systems engineer approaches a complex plant: understand the failure modes first, design for observability, build in human override at every level, and never trust a component you can't evaluate independently.
This isn't just a philosophy — it's how I've built every system in my portfolio.
Interested in working together? I'd love to hear from you.
[email protected]
(667) 355-4069