* Mobile and on-location recording.* Editing and mixing.* Production.

Data and platform engineering; Data science and analytics.

Zymergen is a synthetic biology company that uses genetic engineering guided by data science to deliver end-to-end microbial strain improvement for applications in industrial fermentation. As the founding member and manager of the Data Science team, I worked in both a technical and strategic capacity, establishing the vision and roadmap for our software products and communicating that vision to stakeholders, executives, and our board of directors.

I supported the high-throughput testing platform with analytical automation and applied machine learning techniques to make predictive strain designs.

As a consultant, I developed a 3-course series in applied in deep learning, starting from fundamentals through advanced applications in natural language processing, image recognition, and reinforcement learning. I delivered course outlines, readings, and assignments.As an instructor, I developed and taught a machine learning course for business intelligence. The 10-week, project-based course focused on predictive model development and covers linear and logistic regression, decision trees, association rule mining and clustering.

S3 is a cloud-based storage service that generates immense volumes of data. I was the first member of an embedded research team where I applied a broad set of analytical techniques to inform engineering decisions including segmentation of customer usage patterns, survival analysis to estimate time-to-network-failure, and disk capacity and latency simulation.

The Nordstrom Data Lab is a small team of data scientists and software developers who work on a breadth of data-driven products. We focused on high-risk, potentially high-reward projects meant to disrupt and improve current technologies with the goal of providing Nordstrom shoppers with the best customer experience in retail.

My dissertation research involved using network motifs, small repeating subgraphs that are over- or under-represented in comparison to a random graph, to characterize the metabolism of over twenty organisms from six kingdoms of life: Animalia, Archaea, Bacteria, Fungi, Plantae and Protista. In addition to my thesis work, I was fortunate to collaborate with many scientists inside and outside of my lab and gain exposure to many types of data and analysis methods. I worked largely with high-throughput biological data, specifically untargeted metabolomics, Affymetrix and Illumina microarray, and genome sequence data.

I conducted exploratory data analysis and modeling under the guidance of Drs. James Malley, Lawrence Brody, and Tyra Wolfberg. I concentrated on several projects involving statistical genomics and feature discovery via machine learning. The data consisted of lists of exons from many species and related sequence information (percent identity, amino acid frequencies, etc). We uncovered several novel results using methods of statistical learning machines, specifically Random Forests, and closely related methods for rigorous classification error estimation and model validation. This internship was my first real exposure to advanced statistical methods, and inspired me to pursue a career in modeling and statistics.