I MAKE AI DEFENSBILE.

Clarity at the point of decision.

I work on the part of AI systems that fails quietly: the moment when AI commits to action.

My focus is not whether an AI system works, but on whether it can be explained, challenged, stopped, and defended by the right people at the right time. This means designing clear decision boundaries, human-in-the-loop controls, and traceability into AI systems before they scale, not after they break.

Automation changes who decides, not who’s accountable.

Automation doesn’t remove responsibility, it concentrates it.