Chandler, Arizona, United States
I graduated from Duke University with a Bachelor of Science in Statistical Science and Mathematics, honing programming skills in R, Python, C++, MATLAB, and in statistical machine learning. At my first post-graduation job, I helped develop and test software as a Systems Engineer at Honeywell Aerospace Technologies. Soon thereafter, I accepted a generous fellowship offer from the Computational Science, Engineering, and Mathematics (CSEM) program to start my PhD studies at The Oden Institute for Computational Engineering and Sciences @ The University of Texas at Austin. During my undergraduate studies, I had the privilege of participating in several signature research experiences, including Research in Industrial Projects for Students (RIPS). At RIPS-Singapore, I collaborated with P&G's Singapore Innovation Center to improve their understanding of clinical skin and scalp care trials. At RIPS-Los Angeles, I partnered with the Air Force Research Laboratory to enhance a deconvolution algorithm for physical systems. In both programs, I applied my knowledge of programming and mathematics to develop and implement innovative solutions. At the Math REU hosted by Cornell University, I leveraged ideas from optimal control theory to solve two interesting problems.
SBG: Aerospace Technologies - Engine Systems & Component Analysis - Advanced Technology, Performance, Operability, and Methods Methods team member supporting jet engine test software modernization efforts; Unique opportunity to collaborate with a global software development team
Industrial Mentor: Dr. Pradipta Sarkar My team worked with Proctor & Gamble's Singapore Innovation Center as part of the RIPS-Singapore program. We analyzed the results of clinical skin and scalp care trials using multivariate linear mixed models.
Project Title: Probit Regression Models for Species Occurrence Data in Ecology Advisor: Dr. David Dunson - Literature review of random partition models and Indian buffet-like processes - Implemented Bayesian probit regression for binary vector data - Learned how to adapt/generalize infinite latent feature processes to joint species distribution modeling in ecology
Cohort: Optimality & Uncertainty (Dr. Alexander Vladimirsky) I worked on two projects relating to optimal control theory and piecewise-deterministic Markov processes (PDMPs). In one project, my group developed and implemented mathematical models to study the optimal behavior of prey in a landscape of fear (ecology). In another, we worked on problems in which the agent is generally unaware of the prevailing environment of the process (mode) but can receive sporadic updates about the current mode of the PDMP across the time-horizon. I co-authored a paper describing the results of the first project. My contributions to the second project are acknowledged in a separate publication.
Advisors: Drs. Brett Fischer and Michele Easter In January 2021, I was invited to join a research team at the Wilson Center for Science and Justice (at the Duke University School of Law) as part of its statistical science cohort. I worked on a project where the objective was to document the rates at which criminal cases are dismissed across Virginia counties. Collectively, the three datasets I worked with included approximately 3.3 million observations (cases handled by district courts in Virginia). I cleaned the data and performed preliminary exploratory data analysis in R. I resumed work at the Center in the fall semester and experimented with Automated Case Information System (ACIS) data.