-Developing deep learning models to quantify public opinion-Currently developing natural language interface for large, unstructured data sources with LangChain-Developed various ensemble classification models for natural language processing applications meant to aid government entities in better understanding their audiences and directing campaign strategies-Designed and implemented a software which automatically labeled the sentiment of conversations on an organization's social media page, automating an important task which was previously done manually each day, saving a significant amount of time-Designed and implemented a system to assess public acceptance of a defining policy of an organization-Designed a system to assess public approval for an elected official / democratic organization, allowing an understanding of the most important factors which drive public approval as well as the prediction of public approval in communities external to the database-Wrote two patent documents for own software systems, submitted one which is awaiting approval

-Developed generalized theory of soft contact mechanics for elastohydrodynamic deformation and various disjoining pressures from interfacial physics on arbitrary linear viscoelastic / elastic solids.-Development of nano/micro-rheological measurement techniques for soft, biological materials in fluidic environments using Atomic Force Microscopy (AFM) force-distance (FD) experiments.--Worked with a colleague to develop and implement a viscoelastic characterization technique using the modified Fourier transform which enabled the direct measurement of the complex modulus, something that was not previously with FD experiments--Determined the existence of inherent restrictions in rheological characterization from FD experiments which were then exploited using information theoretic arguments to optimally design experiments to target specific timescales of a material--Derived an expression for the error which results from uncertainty in the AFM probe – surface contact geometry which demonstrates that the physical models to analyze FD experiments are only accurate in overly idealized cases--Wrote a simulation in Python to quantify the pre-contact deformation of a soft, viscoelastic surface by the long-range repulsive interaction with an AFM probe, that the probe – surface contact geometry is never as assumed in the models-Investigation of changes induced by cold atmospheric plasma treatment of cancer cells--Performed in vitro characterization of the viscoelastic properties of cancer cells to demonstrate a potential modality of selectively triggering cell death due to plasma treatment--Further quantifying the degradation of DNA and RNA due to exposure to plasma treatment using AC mode AFM imaging and single molecule force-spectroscopy

-Designed and built several iterations of a micro/milli-scale force microscope intended for the viscoelastic characterization of biological tissues-Wrote unsupervised learning algorithms to automate processing large AFM datasets: labeling contact point, identifying mechanically and topographically distinct regions-Delivered instructional lectures to research group members on topics including applied math, surface science, and machine learning roughly twice per month-Wrote molecular dynamics software from scratch in Python to simulate charge - discharge cycles of a Lithium ion battery. The results qualitatively agreed with physical curves as well as demonstrated anode \ cathode degradation.

-TS/SCI clearance-Worked on the retrofit of COLUMBIA class submarine systems