About

I’m Dallas, a cyber PhD student at the University of Tulsa working on AI Security & Safety. My dissertation builds a trajectory-level framework for detecting behavioral misalignment in AI agents — the idea being that what an agent does over time tells you more about its goals than what it says.

Right now I’m splitting time between three things:

• Dissertation. Alignment Learning Models — extending Decision Transformers with a 5-element behavioral ontology to spot covert objectives.
• Teaching. Delivered CYB-4203/6203 Secure & Trustworthy AI at TU this spring; now building a Red Teaming AI Systems course for Fall 2026.
• Community. Co-founded the TU AI/ML Club and serve as its Outreach Director, after two years as Graduate President.

Before this I was an engineering physicist, then a UAV/robotics product engineer. The thread connecting it all is: build the thing, trace what it does, decide if you trust it.