I build the computational tools that power systems work runs on.

My focus is the gap between what power systems research knows and what practitioners can actually run. I take methods from optimization, operations research, and machine learning and turn them into engines — power flow, contingency analysis, market simulation, network-model tooling — that are fast enough, scalable enough, and dependable enough to carry commercial decisions.

Increasingly, my work sits a layer above the engines: designing decision systems for problems that unfold under uncertainty. That means physics-based scenario simulation as the foundation, stochastic optimization and sequential decision modeling as the architecture, and AI applied where it genuinely earns its place — forecasting the inputs and correcting the residuals — rather than replacing the physics. The goal is the same across markets, operations, and planning: turn a hard, uncertain problem into a repeatable, auditable pipeline that decision-makers can trust.