Provided statistical and analytical support to the CSU Department of Statistics by modeling spatial air pollution data. Analyzed environmental exposures, including PM2.5, Ozone, Light at Night, and NO2, to identify and understand their spatial relationships.

Investigating unsupervised deep learning image denoising methods - specifically Noise2Noise - for use in a CSU microscopy lab pipeline. Developing novel ways to validate and benchmark model predictions without access to ground truth. In collaboration with Dr. Jesse Wilson and the Walter Scott Jr. College of Engineering.- Created a fully operational model that instantly increases PSNR of highly corrupted images by 200% by utilizing HPC, PyTorch and Weights & Biases to train on large-scale image data.- Formulated a novel way of verifying the results of the Noise2Noise framework. Currently in the process of publishing a manuscript on our findings.

Tutor in the Calculus Center at CSU and TA for Calculus I.

Analyzing effectiveness of Thin Plate Regression Splines for controlling unobserved spatial confounding. Extending traditional work from linear, continuous data to Poisson distributed count data. Overseen by Dr. Kayleigh Keller and the Department of Statistics at CSU.