Work experience summary:Pennsylvania State Postdoc: Aug 2019 – September 2023• Led an interdisciplinary team of astronomers and NASA data scientists to pioneer methods using Deep Learning (DL) to improve our ability to measure exoplanet masses by modeling non-linear features in noisy data• Develop automated pipeline in Python and Julia to prepare data for input into DL code segment• Interpret DL results and process them to be digestible and convincing to the exoplanet mass community• Develop Julia package, RvSpectML.EchelleCCFs, for statistical methods to reduce data for planet mass measurement• EchelleCCFs was selected to reduce data for international data challenge because of high-fidelity data products• Develop Julia package, RvSpectML.RvLineList, for generating data products to aid planetary mass data analysisUniversity of Delaware Graduate Student Aug 2013 – Aug 2019• Translate S.O.A.P., a program written in C for simulating stellar spectra, to python• Invent and code in python an automated method to normalize spectra more accurately than previous manual methods• Design and write new python code to analyze large (100 GB) archival dataset of stellar spectra• Visually explore dataset using python plots and movies to find new stellar noise indicators in these datasets• Exploit correlations with known noise indicators to discover new stellar noise indicators in automated search• Modify FARGO, a C-based code simulating planet-forming and migration, to support new accretion physics• Set up and carry out planet simulations in Linux-based high-performance computing (HPC) environment• Write custom Bash scripts to automatically submit HPC jobs and manage output files• Taught self to use Python (scipy, numpy, pyplot, matplotlib) to analyze HPC simulation results