February 14, 2026 - From Watching Demos to Building Tools: My AI Journey in Emergency Management About a year ago, I sat in the CISA AI and Machine Learning Bootcamp learning about models, algorithms, and frameworks. We’d type in code snippets to “see what this does,” and while informative, something was missing. I kept thinking: I see what’s possible. Now how do I actually build something for chemical emergency response?

That question wouldn’t let go.

I’ve spent two decades working CBRN threats from every angle—operations, policy, preparedness. I could see how AI could transform our training and response capabilities, but attending lectures wasn’t going to get me there. So I stopped being a spectator and became a student again.

I enrolled in Udemy and Coursera courses. Took data analysis courses. Learned system design. And honestly? It’s been humbling. The hardest part hasn’t been the concepts—I think in systems, so understanding how pieces connect makes sense to me. The challenge has been the language. Developers name things in ways that aren’t remotely descriptive, and figuring out where each piece goes in the architecture feels like solving a giant jigsaw puzzle without the box cover—and all the pieces are upside down.

But I’m building anyway.

Right now, I’m working on a web-based chemical incident simulation platform that uses AI to generate dynamic scenario injects and model crowd behavior—the unpredictable human element that makes or breaks real-world CBRN response. The goal is to bridge the gap between tabletop exercises (which lack realism) and functional exercises (which are resource-intensive) by creating stress-test scenarios that responders can run without pulling entire teams off the line.

Is it perfect? Not yet. Am I still Googling basic terminology? Absolutely. But every day I’m getting closer to building tools that can actually help our field prepare for low-frequency, high-consequence threats.

Why share this? Because if you’re in emergency management, public health, or any field where you see AI’s potential but don’t know where to start: you don’t need to become a data scientist overnight. You just need to start building. Pick one problem you understand deeply. Find one tool that might help solve it. Break things. Ask questions. Repeat.

If you want to talk about the problem you want to solve, reach out! Happy to work with you.

The future of our field isn’t just using AI tools others build for us—it’s building tools shaped by the expertise we’ve spent careers developing.

Courses and resources that helped me get started: DHS/CISA’s AI and Machine Learning pathway taught by Eylem Uysal, Angela Yu’s Python course on Udemy, Ed Donner’s LLM and Agents courses on Udemy, and a bunch of IBM, AWS, and Azure courses on Coursera.