- Collaborated with Dr. Timothy Menzies on a project to create a predictive model for the health of computational science projects in the future.- Automated the scraping of indicators from 150+ GitHub projects.- Utilized various machine learning models, including CART and KNN, to predict project health indicators.- Conducted statistical tests to assess the accuracy of the predictive models.- Performed an extensive literature review and documented the methodologies and findings for the project.

- Spearheaded the design and execution of a resilient continuous integration pipeline for Randles lab using Gitlab CI.- Demonstrated significant achievements by enhancing the efficiency of the pre-existing pipeline by 83% through meticulous pipeline reorganization.- Seamlessly integrated high-performance computing clusters into pipelines to facilitate the execution of compute-intensive tasks across multiple nodes.- Expedited and streamlined the setup of dependencies by leveraging Docker and Gitlab Container Registry for efficient containerization.

- Orchestrated the development of a SQL and Python-driven data migration pipeline to transition data to GCP BigQuery, ensuring data robustness and integrity.- Enhanced user experience by simplifying schema modifications through the integration of Google Sheets and Cloud Functions as user-friendly interfaces.- Implemented efficient data models and aggregations to facilitate the creation of tailored Looker dashboards.- Achieved a 70% reduction in dashboard rendering time by automating datamart generation through optimized SQL queries.- Oversaw the deployment and upkeep of artifacts across various environments using version control systems like Git and specialized tools such as Gazer.- Provided leadership and mentorship to a team of 8 individuals, guiding them on tools and project execution.

- Collaborated with a diverse group of developers in the open-source community "OpenMined", dedicated to advancing privacy-preserving machine learning models.- Made contributions to the Pysyft library, focusing on the implementation of secure multiparty computation for key neural network layers, including Dropout and Dense layers.