Anushka Murthy

Anushka Murthy

PhD Student at Stanford University

About Me

I am a PhD student at Stanford where I’m very fortunate to be advised by Ramesh Johari and Irene Lo, and to work closely with Gabriel Weintraub. I grew up in Michigan and completed my B.S. in Applied Math and CS at Columbia University. Up until the beginning of grad school I was primarily interested in Probability Theory and Theoretical CS, but now I am mostly interested in using techniques from these areas to study problems at the intersection of Market Design and Data Science. At Stanford, I am affiliated with the Operations Research group in the MS&E department, the Data Science Center, and the Society and Algorithms Lab. I’m also thankful to be supported by a Stanford Data Science Scholarship, NSF Graduate Fellowship, and an EDGE Doctoral Fellowship.

News

  • [June 2025]: Starting my first ever internship! I’m joining Amazon’s Middle Mile Product and Technology team in Bellevue as an Applied Scientist intern. Looking forward to working on ML and experimentation methods for the freight marketplace!
  • [May 2025]: Giving three talks on our paper, “When Does Interference Matter? Decision-Making in Platform Experiments.”
  • [October 2024] : Giving three talks on a subset of results from our paper, “When Does Interference Matter? Decision-Making in Platform Experiments.” The talk will be under the title, “False Positives in A/A Experiments with Correlated Outcomes.”
  • [September 2024]: Excited to join the 2024-2026 cohort of the Stanford Data Science Scholars Program!
  • [July 2024]: I will be giving a talk at MSOM 2024 at the University of Minnesota on our “Dynamic Contracting for PES Programs” paper.
  • [June 2023]: Excited to TA the “Mathematics and Computer Science of Market Design” summer graduate school at MSRI!

Research Interests

A consistent theme of my research is the design and evaluation of dynamic interventions to guide learning and decision-making in markets. I typically use techniques from queueing theory and dynamic optimization to study two different sorts of questions: (1) how should market policies be designed for resource-constrained settings? And, (2) how should the data science methods used to evaluate these policies be designed? I enjoy working on applications with a social good impact, and I’ve been particularly interested in sustainability and education technology. Two distinct (but not necessarily disjoint) themes in my research are:

  • Experimentation and Measurement in Markets: I am excited about using stochastic modeling of marketplace dynamics to design causal inference methods that can handle dynamic interference and dynamic treatment effects. So far, I have worked on A/B testing in online marketplaces, with a focus on how interference affects decision-making from a hypothesis testing perspective. I’m currently focusing on personalized learning platforms and studying dynamic experimentation under student learning dynamics. This has led me to interact with Ben Domingue and his fantastic group in the Graduate School of Education, and I’ve also been doing some domain-specific work on the measurement and psychometric modeling of learning dynamics.
  • Market Design for Social Good: I am also excited about using game theoretic modeling to study public sector markets that deal with severe information asymmetries and resource constraints, where it is crucial to design policies that are simple, informationally robust, and equitable. So far, I’ve studied Payment for Ecosystem Services (PES) programs and worked on simple yet approximately optimal dynamic contract design. I’m currently studying the design of recommender systems and AI-driven information platforms in school choice to help applicants learn their preferences.