About Stephan Claxton
I design and architect engineered systems for the era of AI and autonomy. The kind where software, hardware, machine-learned components, and human operations all have to behave as one coherent whole. My work sits at the intersection of model-based systems engineering and the practical realities of shipping high-consequence software that has to be right in environments you can’t fully predict.
Most of my career has been spent on systems that don’t tolerate ambiguity: autonomous vehicles, spacecraft, and defense platforms. The common thread isn’t the domain. It’s the discipline of treating architecture as the actual design of the system, not a diagram you make after the code is written.
Today that work is on autonomy at Applied Intuition, doing systems engineering for AI-driven vehicle development. Before that, I led senior systems engineering and technical program management at Strategic Technology Consulting (an Arcfield company), and I started in aerospace at Lockheed Martin doing model-based systems engineering for space and defense programs. That’s where I learned to take models seriously, not as compliance artifacts but as the design of systems you can’t reach once they’re deployed.
Things I care about
What systems engineering becomes in the age of AI and autonomy. Executable models and what they mean for verification when the operational design domain is open-ended. Requirements traceability as a survival skill, not a checkbox. Architecture as the real product of systems engineering. The cultural translation problem of bringing aerospace-grade discipline into software-cadence development without slowing the whole thing down.
How I think about this work
The intellectual anchor for all of this is the systems-thinking tradition, Donella Meadows in particular, alongside Nancy Leveson’s work on safety in complex software-intensive systems. The discipline asks not just what is this system supposed to do but what behavior is its structure producing, and which interventions would actually change it. MBSE, SysML, and executable models are tools. The discipline underneath is systems thinking applied seriously to engineering work that has to be right.
The thing Meadows got right that the engineering world is still catching up to: structure produces behavior. The fastest way to change what a complex system does is to change the structure that produces its behavior. Not the people running it, not the metrics on top of it, and not the language used to describe it. Most of what I write is some version of that observation, applied to the engineering of autonomous and AI-driven systems.
Education
MBA in Finance from Embry-Riddle Aeronautical University. Bachelor of Engineering from Western Michigan University.
Get in touch
The best way to reach me is email or LinkedIn. I’m especially interested in conversations with founders and technical leaders building in AI, autonomy, and other high-consequence software domains, as well as with engineers thinking about what good systems-engineering practice looks like at startup cadence.