Assistant Data Scientist
Raleigh-Durham, North Carolina Area
- Supporting developer of rGenesis, an R Shiny forecasting application for simulating recruitment in clinical trials.Utilize Bayesian techniques with Gamma-Poisson, Beta-Binomial and Weibull models.Increase efficiency of simulation using data.table and parallel processing.Implement intuitive UI with dynamic data visualizations.Connect to multiple data sources – databases, input csv files, APIs.- Junior developer of the Protocol Design Tool, an R Shiny application that analyzes the impact of inclusion and exclusion criteria on the available patient population for a clinical trial.Ingest and visualize de-identified patient data.Implement data validation on ICD-10 codes using LabCorp test data.Construct query builder for Citeline API to compare protocols for a given indication, phase, etc.Develop downloadable, custom reports in HTML, RTF or PDF based off exploratory analysis.Visualize variable importance from random forest model to determine importance of inclusion/exclusion criteria.- Converted an XML to PostgreSQL ETL pipeline to an API to PostgreSQL pipeline.