Research

My research focuses on sequential decision problems under uncertainty and data-driven decision making. I use stochastic modeling and techniques from optimization, applied probability, and statistics to address challenging operational problems with complex uncertainties and tradeoffs. I develop practical and efficient algorithms, and derive provably strong guarantees for their performance. In past and ongoing research, I have worked on problems in sequential search, pricing, scheduling, and optimal stopping, as well as problems in dynamic resource allocation and revenue management.

I am particularly interested in sequential search, a classical problem at the intersection of CS, Econ, and OR. Sequential search problems, such as Weitzman's Pandora’s problem, involve sequentially exploring a given set of alternatives when exploration is costly. With applications across many areas — spanning consumer search, hiring pipelines, new product development, and training ML models — sequential search problems are highly relevant to a variety of decision makers, including individuals, businesses, and AI agents.