San Francisco Bay Area
• Conducted independent research project titled “Evaluating Machine Learning Methods for Event Classification in the Active-Target Time Projection Chamber” building on work for the National Superconducting Cyclotron Laboratory in summer 2017. • Applied machine learning methods to nuclear scattering data and evaluated their viability as part of the analysis process. • Integrated a new step into the existing detector data analysis framework to increase statistical power of experimental results. • Used Python libraries including NumPy, pandas, h5py, scikit-learn, and Keras.
• Built a multi-output, multi-class image recognition model in Python using Keras to be integrated into the company platform. • Utilized a pre-trained network architecture to classify and tag product images based on a set of attributes using a small training set.
• Worked as a Research Associate/Technical Aide for the National Superconducting Cyclotron Laboratory (NSCL) at Michigan State University. • Documented a Python data analysis package for analyzing nuclear scattering data produced by the lab’s Active-Target Time Projection Chamber. Created an analysis manual for the software suite to improve analysis steps and aid future researchers. • Cleaned, fit, and analyzed data from a 40Ar beam experiment from raw binary files to compressed HDF5 format. • Selected to present work at the Fall Meeting of the Division of Nuclear Physics in Pittsburgh, PA - October 2017.