Research

The convergence of artificial intelligence, energy systems, and sustainability presents unprecedented opportunities and challenges for society. My research addresses a fundamental question: How can organizations make better decisions about energy investments and operations in an uncertain and resource-constrained world? At the core of my work is the development of computational methods that automate complex decision-making processes, making sophisticated optimization tools accessible to practitioners while advancing the theoretical foundations of the field. My research program operates at the intersection of large-scale computing and energy. I develop self-adapting approximations of Markov Decision Processes (MDPs), which are the foundational models underpinning reinforcement learning. These methods automate critical steps in model selection and solution guidance. They have been applied to manage production and storage energy real options, and to tackle emerging procurement and generation/transmission capacity problems.