Experience

For more details, please see my CV.

Edwards Lifesciences

Consultant: Machine Learning (June 2022 - Present) – Irvine, CA

• Using clinical trial and real-world data to extract insights and build predictive models for several use cases including patient retention, adverse event prevention, indication expansion, product design, and precision medicine.

• Frequent collaboration with several different functions within Edwards such as engineering, medical affairs, biostatistics, and operations.

• Investigated the use of machine learning-generated synthetic patient data to improve predictive model performance and statistical inference for potential gains in trial diversity and cost savings. The work was presented and published at Computing in Cardiology 2023 (IEEE).

Machine Learning Intern (June 2021 - Sept. 2021) – Irvine, CA

Worked in the Transcatheter Aortic Valve Replacement (TAVR) Global Data Science Team developing interpretable machine learning methods for cardiology applications.

• Conceived, developed, and pitched to senior leadership a machine learning patient matching algorithm and dashboard for TAVR precision medicine. The product was a top 5 finalist in a company-wide “shark tank” competition and was presented to the senior executives including the CEO.

• Conceived and developed an unsupervised clustering framework for patient phenotyping related to adverse events using R and Python. The work was presented and published at Computing in Cardiology 2022 (IEEE).

Tools: Python, R

University of Southern California

Research Assistant, Data Science (March 2019 - June 2022) – Los Angeles, CA

Consulted with the USC Department of Health Economics, Schaeffer Center, and the Keck School of Medicine applying machine learning and causal inference to explore and evaluate potential treatments for cardiovascular and respiratory disease.

• Developed a statistical analysis plan for an NIH R21 grant application for exploring the use of inotropic infusions in end-stage heart failure patients using real-world data and machine learning.

• Led an analysis using machine learning-assisted instrumental variable analysis with random forests and LASSO regression to evaluate end-stage heart failure health outcomes in real-world evidence (using R). Prepared data via merging and cleaning large insurance claim and electronic health records datasets with billions of rows using SAS and SQL.

• Conducted statistical analysis (survival analysis) and writing for a paper published in the International Journal of Cardiology.

Tools: R, Python, SQL, SAS

Stanford University

Research Assistant, Data Science – Priest Lab (May 2019 - Jun 2021) – Palo Alto, CA

Collaborating with cardiologist Dr. James Priest, M.D. to analyze and model the impacts of cardiovascular disease and related surgery on pulmonary and neurological function

Paper published in leading cardiovascular journal JAHA (second author) and another manuscript in development (first author)

Tools: R, Python (Sci-kit Learn, Pandas)

Research Assistant, Data Science – QSU (Jun 2018 - Jun 2021) – Palo Alto, CA

Consulting with clinicians from the School of Medicine through the Quantitative Sciences Unit (QSU) working on multiple projects and manuscripts across various medical disciplines

Tools: Python (Sci-kit Learn, Pandas, Matplotlib), R

MarketPsych

Data Science Intern (March 2018 - April 2019) – San Luis Obispo, CA

Research into cryptocurrency trading strategies for a finance sentiment data aggregation startup.

• Quantitative analysis of BTC trading patterns based on news and social media NLP using Matplotlib, Pandas, and Numpy.
• Development of sentiment analysis-based bitcoin (BTC) trading algorithms using random forests

Glynt.ai

Data Science Intern (March 2017 - Sep 2017) – Mountain View, CA

Data cleaning, validation, and analysis for SaaS startup that aggregates utility consumer data and trends with Python.