My work focuses on using gameplay telemetry to better understand a game’s player population and the player experience, supporting a variety of titles including DOOM Eternal, Deathloop, and Hi-Fi Rush. In addition to supporting game studios, I work with our corporate marketing team to evaluate the effectiveness of paid campaigns. Using Python and PySpark, I utilize statistical and machine learning techniques including regression analysis, clustering, and skill rating systems such as Glicko-2. Beyond conducting analyses, I collaborate with game design and marketing teams to put these findings into action to improve the player experience and marketing decision making, respectively.

My work focuses on using gameplay telemetry to better understand a game’s player population and the player experience, supporting a variety of titles including DOOM Eternal, Deathloop, and Hi-Fi Rush. In addition to supporting game studios, I work with our corporate marketing team to evaluate the effectiveness of paid campaigns. Using Python and PySpark, I utilize statistical and machine learning techniques including regression analysis, clustering, and skill rating systems such as Glicko-2. Beyond conducting analyses, I collaborate with game design and marketing teams to put these findings into action to improve the player experience and marketing decision making, respectively.

My work focuses on using gameplay telemetry to better understand a game’s player population and the player experience, supporting a variety of titles including DOOM Eternal and Deathloop. In addition to supporting game studios, I work with our corporate marketing team to evaluate the effectiveness of paid campaigns. Using Python and PySpark, I utilize statistical and machine learning techniques including regression analysis, clustering, and skill rating systems such as Glicko-2. Beyond conducting analyses, I collaborate with game design and marketing teams to put these findings into action to improve the player experience and marketing decision making, respectively.

I conceptualized and created autoregressive influenza forecasting models in R for participation in the Centers for Disease Control and Prevention's prospective weekly influenza forecasting challenges during the 2018/19 and 2019/20 influenza seasons. My model placed 10th out of 40 teams during the 2018/19 challenge (2019/20 was not scored due to the COVID-19 pandemic). Additionally, I participated in the FluSight Network, a collaboration of top influenza forecasters working to improve ensemble approaches to influenza forecasting, where Protea models were some of the highest-performing models contributed.

I provided statistical support for a variety of researchers in the MRC/Wits Developmental Pathways and Health Research Unit, assisting in the selection of analytical techniques, appropriate imputation of missing data, and utilization of statistical software on projects.As a member of the DPHRU data team, I co-managed the migration of all collected study data from the Birth to Twenty Plus cohort study (comprising 18 study visits over almost 20 years) into REDCap from a variety of Stata, SPSS, Excel, and Access files.

I developed and executed an analysis plan to investigate and classify patterns of longitudinal changes in persistent environment contaminants among individuals in the SEARCH for Diabetes Youth Study. Building on this, I developed and executed an analysis plan to estimate the health effects of identified categories of environmental contaminant exposure using Bayesian kernel machine regression.Additionally, I analyzed longitudinal data investigating associations of national-level measures of development and health care with outcomes of BMI and overweight percentage for the Lancet Commission on Obesity

I co-managed the CDC's annual seasonal influenza forecasting challenge, consisting of over twenty teams from around the globe. I created forecast visualizations in R for public display and evaluated forecast accuracy at the conclusion of the season. Additionally, I developed guidelines for the inaugural challenges to forecast influenza-associated hospitalizations and state-level influenza activity, both of which began in the 2017-2018 influenza season. In addition to my work on forecasting, I used data from the New Vaccine Surveillance Network to estimate pediatric influenza vaccine effectiveness to prevent influenza-associated hospitalizations and investigated causes of death in patients dying within 30 days following an influenza-related hospitalization using data from the Influenza Hospitalization Surveillance Network.

Using survey response data from the Gulf Long-Term Follow-Up Study, a cohort of responders to the 2010 Deepwater Horizon oil spill, I assessed associations between exposures to oil dispersants used in spill clean-up and adverse respiratory, dermal, and eye irritation effects.

I co-managed the harmonization of SAS datasets for a large-scale meta-analysis consisting of pooled data from eighteen cohort studies in the NCI Cohort Consortium totaling over 900,000 participants in collaboration with Institute for Cancer Research in London. Additionally, I contributed to the development of an analysis plan to examine associations between pregnancy-related measures and premenopausal breast cancer.

I conducted data cleaning and ran basic analyses on a complex data set involving 500+ participants for a validation study examining a home test kit for HPV.

As a trip leader, I managed a team of 5 other instructors to ensure all food, gear, medical, and transportation related logistics ran smoothly throughout the trip (2013, 2014). As a team, we introduced students to and supervised outdoor activities such as backpacking, rock climbing, whitewater rafting, canoeing, mountain biking, and mountaineering.

I taught sixth grade Earth Science (2010-2014) and seventh grade Earth Science (2009-2010), where I refined a demanding, inquiry-based curriculum based on principles of geology, meteorology, and astronomy. Prior to the 2012-2013 school year, I transitioned Earth Science to a “flipped classroom” model, utilizing technology to deliver lectures to students at home while maximizing collaborative time in the classroom. Throughout, I utilized student performance data to customize lesson plans and review activities for particular class periods and individual students

Designed and led practices for 40 middle school wrestlers.

As a Teach For America corps member, I taught Biology (2007-2008) and Environmental Science (2008-2009), during which I led my 11th grade Environmental Science students to an 82% passing rate on the Georgia High School Graduation Test.