Nathan Brouwer

About Nathan Brouwer

Data Scientist and educator with 15 years of experience using Python and R to extract insights from complex datasets. I have extensive experience with all major statistical analyses, supervised learning algorithms, and unsupervised learning methods, including linear and logistic regression, random forests, neural networks, and k-means and hierarchical clustering. I use R and Python daily to process data, make actionable predictions, and create compelling graphics to reveal complex patterns. I just concluded my last semester as a Teaching Professor at the University of Pittsburgh, where I held teaching assignments focused on quantitative methods, computational biology and introductory biology. As an instructor for large introductory courses, project management and science communication were key to my success. Each semester I ran one to two large courses and had up to 300 students. While meeting multiple deadlines to deliver lectures, assignments, readings, and exams I also developed and delivered engaging in-person and online content. This fast-paced environment required me to play the simultaneous roles of writer, producer, director, web-designer, and data analyst. This resulted in a multi-tasked cycle of planning, working, and evaluating to render multiple deliverables each week (lectures, assignments, exams, etc.). I have finely honed my ability to set priorities, manage competing tasks, track progress, and organize project materials.In all aspects of my work I take a data-driven, computing-based approach. Whenever possible I develop reproducible workflows to develop and deliver content to my research collaborators and students. Examples of my hybrid data science-teacher approach include: writing and updating my syllabi with scripts, extracting data from the primary literature to develop engaging new visualizations, using R to manage grades and information on student engagement, and writing software to build and manage exams. My academic research and consulting primarily involve the analysis of complicated longitudinal data to characterize ecological trends. This requires careful development of both data cleaning and quality control workflows and carefully designed statistical models. My commitment to creating high-quality and validated data goes back to my first semester of my PhD work. As a first-year student, I was instructed to hand-curate a decade-long dataset, and I quickly realized the need to do tasks like this computationally with scripts. I\'m now passionate about fully documented workflows, quality control, and data curation.

Experience

1. Senior Data Scientist

   Highmark Health

   Sep 2024 — Present · Pittsburgh, PA, US

Education

• Michigan State University

  Ecology

  2007 — 2008

• University of Pittsburgh

  Doctor of Philosophy (Ph.D.), Ecology and Evolutionary Biology

  2009 — 2015

• Seattle Pacific University

  Bachelor of Science - BS, Biology, General