Software Tools for Green Fleet Management

August 2021 - Present

Designed and developed a web-based software application to manage and track user vehicle assets and fleets using Python Flask. Integrated tracking of emissions and costs to evaluate the potential environmental and emission impact within the management tool. Implemented an optimization model developed to recommend fleet management decisions to reduce emissions and meet emission reduction requirements while minimizing costs.

October 2021 - December 2021

Researched the current state of generative adversarial network (GAN) models. Implemented a cGAN model on a subset of Google’s Quick! Draw dataset to generate synthetic computer drawn doodles. Trained cGAN model with different optimizers (SGD, Adam, RMSProp, etc.) to determine best optimization method, analyzing with statistical and hardware measures. Compiled research and experimental findings in a formal ICML research paper.

November 2021 - December 2021

Implemented machine learning and data analytics methods (neural networks, SVM, PCA) to predict and analyze tumor cell malignancy in the Wisconsin Breast Cancer Dataset with up to 98% test accuracy. Reported the viability of using these methods to predict the malignancy of cells in a cumulative report discussing the background and implications of our findings.

Pong & 3D Breakout

August 2021 - December 2021

Recreated the classic game of Pong as well as extended it to a 3D version of Breakout in Javascript with the addition of increasingly difficult levels and limited lives. Implemented game design elements and animation principles such as squash and stretch and secondary action using computer graphics methods such as transformation matrices and shaders.

September 2020 - December 2020

Built a database management system in Java featuring a SQL parser, the capability to index data for fast access, multiple implementations of relational operators, and a query optimizer. Extended management system to offer parallel implementations of various relational operators to support faster queries. Conduct experiments on various relational operators to determine the optimal operator for various queries.

June 2020 - August 2020

Implemented transfer learning for image classification on spotlight SAR imaging from MSTARS dataset in Python with up to 97% test accuracy. Performed experiments on different transfer learning approaches to determine viable transfer learning techniques for SAR imaging. Performed analysis on transfer learning approaches to determine the viability of using transfer learning techniques versus standard machine learning techniques.

May 2018 - July 2018

Designed and built a system to map photovoltaic response of solar cells based on laser beam induced current (LBIC) to analyze active areas, defects, inactive regions, and coating errors to inform on how to improve production solar cells and provide images and data on overall photovoltaic response. Converted laser engraver to low-power scanning laser to scan solar cells and collect photovoltaic response in CSV files to be mapped into images of photovoltaic active areas of a solar cell. Developed a MatLab program to extract data from a scanning laser device to map the photovoltaic response from induced currents.