Adam Shelton Email and Phone Number

As Data Manager, I have supported professional staff and researchers at the Center for Neighborhood Engaged Research & Science at Northwestern University. I have completed a wide range of data and infrastructure related projects including:- Transitioned quantitative data storage from on-premises file server to Google Cloud BigQuery data warehouse- Developed automated Python/SQL cloud ELT pipelines for retrieving/processing data with serverless cloud functions- Employed DevOps principles by creating Cloud Build continuous integration (CI/CD) pipelines to deploy infrastructure and build container and Packer VM images from configuration as code stored on GitHub- Improved data utility by integrating Google Maps geocoding API and Document AI API calls into data pipelines- Developed R and Python packages to standardize data visualizations/documentation and improve existing code usability- Created a centralized documentation system and Data Studio dashboards to promote data democratization- Tested various orchestration platforms and devised a transition plan of data pipelines to Apache Airflow

I coordinated with educational researchers in the company and school district to improve student outcomes through a variety of quantitative methods. I built Random Forest and Support Vector Machine models using machine learning methodologies to predict the adoption of specific computer science curriculum. I also conducted factor, principal component, and unsupervised cluster analyses to find patterns in student survey data and composed written reports and presentations to convey the methodologies and findings of my research. Recently, I independently developed an open-source course on R for the company and taught it to post-doc researchers over the course of a few sessions. I am currently co-authoring two papers on predicting Computer Science course outcomes using Hierarchical Linear Modeling.

I collaborated with faculty in the Sociology department, conducting social network analysis, investigating ties between demographics, relationship dynamics, and health. I organized and cleaned datasets using R to prepare data for analysis, created new covariates and recoded existing variables using R to aid in data modeling, and generated initial bi-variate models utilizing linear and logistic modeling methods in R.

I worked a summer in the Roll Out Center of CVS Health, providing phone support as well as data acquisition and analysis to coordinate a wide variety of projects with pharmacies, contractors, and local governments nationwide.