Managed and analyzed over 250 million records from prescription datasets, such as CURES and IQVIA, using Python and SQL to extract, transform, and load data for assessing opioid concentration in California. I applied advanced statistical techniques, including hypothesis testing, regression, classification, and time series analysis with R and Python, to create data visualizations that highlighted addiction patterns for reporting. Through exploratory data analysis (EDA), I identified key trends and anomalies, presenting clear and concise recommendations to non-technical stakeholders. Additionally, I streamlined data processing by creating tailored DataFrames, which reduced SQL query execution times by 45% and improved ML model training efficiency by 30%.

Played a pivotal role in improving the internal testing infrastructure for the SAAS product Spinnaker. This involved creating new testing pipelines, allowing us to effectively test third-party software like Open Policy Agent (OPA). In addition, I designed and implemented Jenkins test pipelines that seamlessly coordinated with Spinnaker services, providing clients with a clear view of Spinnaker's integration capabilities. I also optimized the performance and efficiency of our testing process by configuring Prometheus, a metric monitoring service, and integrating it with an Observability plugin. This effort led to a significant reduction in pipeline execution times. To streamline automated monitoring of pod deployments, I crafted shell scripts to collect crucial data from various namespaces in EKS clusters. Lastly, I kickstarted a resource efficiency project by deploying Jenkins and Prometheus as durable services within individual pods in an EKS cluster, resulting in a reduction in resource creation times.