Developing and deploying scalable predictive modeling systems for a variety of applications in the credit space.

- Designed and built scalable machine learning solutions for the model development and validation process.- Built predictive models from large (TB scale), highly multivariate credit card data sets- Built and maintained validation and visualization tool kits in response to and in anticipation of business needs- Frequently taught and lead training sessions for data scientists on distributed infrastructure and scalable machine learning- Engaged with engineering and business teams to prepare robust and well-validated solutions for our stakeholders

Contractor supporting the Department of Homeland Security/Customs and Border Protection for radiation monitoring and nuclear threat detection program. I Wrote analytical tools to provide quantitative insights on data collected by a nationwide fleet of Radiation Portal Monitors deployed at all points of entry into the US and performed ad hoc analyses for reports as needed.

Thesis research in observational astrophysics. Using optical/near-UV spectra obtained at major ground-based telescopes, we performed high-precision measurements of the primordial Deuterium/Hydrogen abundance ratio towards quasars. This abundance ratio acts as an important constraint on models of the early universe (roughly 10 seconds to 20 minutes after the Big Bang). A second project used a sample of ~10000 absorption systems to characterize the thermal evolution of intergalactic gas 10-12 Gyr ago. Wrote a software framework for analyzing and fitting quasar absorption lines.Wrote tool to constrain absorption models and estimate confidence intervals on fitting parameters using stochastic optimization methods.

Teaching assistant for a wide range courses for both technical and non-technical undergraduates and laboratory and lecture settings.

Worked on quasar absorption spectroscopy.

Summer research project in solid state physics.