▪ Developing a Conditional Generative Adversarial Network (cGAN) in order to identify the crystal structures of theoretical materials based on desired physical properties▪ Created binary, ternary, and pseudo-binary phase diagrams with Thermo-Calc to examine ideal conditions capable of sustaining BSTS (BiSbTeSe2) single crystal growth to assist in the study of topological quantum computing

▪ Research and development in the field of low Reynolds number fluid dynamics and biomechanics▪ Team lead in the construction of a haptic simulator to emulate the movement of microbials▪ SOLIDWORKS and Arduino were used for design and testing

▪ Created software pipeline for generating synthetic semantic parsing training data for Seq2Seq models representing Google Assistant in low-data environments using Google’s Transformer-based large language model, LaMDA▪ Using 15 showcase examples or less per dataset, constrained decoding, and LaMDA’s few-shot learning capabilities, successfully generated 20 full semantic parsing training sets comprising over 300,000 examples

▪ Documented preliminary testing results for a new motor initiation system that will be used in NASA’s Space LaunchSystem (SLS) on Artemis III▪ Performed statistical analysis on change report life cycles to identify systemic departmental issues that preventefficient distribution of information▪ Investigating design alternatives for the SLS Aft Exit Cone Shipping Container to ensure reliability and safety

▪ Researched the capabilities and limitations of implementing machine learning in particle accelerator experiments▪ Utilized electron beam data from previous plasma wakefield acceleration simulations to create optimization systemscapable of locating an electron beam’s current profile and longitudinal phase space using neural networks created inPython with TensorFlow▪ Reduced convergence runtime from several hours in previous optimization attempts to approximately 0.1 seconds