GA Tech

Introduction

The Fall of 2021 marked the beginning of my pursuit toward personal advancement as I enrolled in Georgia Tech's Professional Education program to earn my Master's in Electrical and Computer Engineering. This degree in Electrical and Computer Engineering (ECE) seamlessly integrates coursework from both disciplines, enabling me to explore a diverse range of topics. By taking a total of 10 courses across these fields, I had the flexibility to address gaps in my knowledge from my undergraduate majors of Physics and Mathematics while also gaining experience in areas I sought to grow professionally. The Professional Education program provided the flexibility and support necessary to maintain my commitment to academic excellence at Georgia Tech while seamlessly managing the responsibilities of my full-time role as a software engineer. As a testament to my dedication and the quality of this program, I was able to graduate with a 4.0 GPA while during this time becoming an integrated product team lead for three actively deployed technologies, managing tasking for a team of nine engineers and scientists, and achieving recognition as an inventor for two awarded U.S. patents.

In my undergraduate education at the University of Florida, between my Physics labs and Mathematics classes, I was able to take a few courses that covered computer engineering and software development such as Intro to C, Numerical Analysis, and Physics Modeling in Python. I was also able to use MATLAB intensively within my undergraduate research role for 2 years and Mathematica while guest lecturing for a 4000-level Mechanics 2 course. While these ultimately became the skills that got me my first job in software engineering, I felt as though there was still much more that I could learn to become even more proficient.

My education at Georgia Tech profoundly shaped my understanding of both Electrical and Computer Engineering, equipping me with advanced skills that significantly elevated my career. The program's flexible cource selection allowed me to strategically tailor my studies to complement the cutting-edge technologies I actively researched and developed at work. Through this journey, I deepened my expertise in computer architecture and software development, mastered advanced mathematical concepts such as Fourier analysis, signal processing, and partial differential equations, and ventured into new domains like computer network security and adaptive filtering.

Continue reading to discover a showcase of impactful projects I developed during my time at Georgia Tech along with an in-depth look at the coursework that shaped my advanced education in Electrical and Computer Engineering.

Project Highlights

Network Security for Websites in China

Video Presentation
Join me in this brief video presentation as I delve into the what I learned in querying over one million websites to analyze and extract critical insights about network security.

Academic Paper
This paper explores the tools, methodologies, results, and conclusions of this study — offering valuable insights and a comprehensive examination of the findings. Give it a look!

Summary

This project was inspired by the discovery of Why No HTTPS, a website that catalogs popular platforms delivering content over unencrypted connections. A substantial number of these websites are hosted in China, raising questions about the state of secure web protocols in the region.

The goal of this research was to examine the adoption and implementation of HTTPS and SSL certificates in online services based in China. By analyzing trends in secure web traffic, this project explored the broader implications of utilizing—or neglecting—the TLS protocol, affecting both servers and clients interacting with these services.

For a deeper dive into prior studies, methodology, results, and conclusions, refer to the video presentation and academic paper shown above. Additional resources are available below, with further insights into this Computer Network Security course outlined in the Coursework section of this page.

See More From This Project
Video Presentation
Academic Paper
Presentation Slide Deck

Modeling and Simulation Ticket to Ride

Video Presentation
My incredible teammate, Rachel, demonstrates the capabilities of our Ticket to Ride Simulation program in MATLAB, showcasing the precision and innovation that make this project truly stand out.

Academic Paper
This academic paper, written collaboratively by my teammates and me, provides an in-depth exploration of our innovative approach to developing this versatile simulation tool.

Summary

This group project provided a unique opportunity to explore the intersection of simulation and game theory through the popular board game Ticket to Ride. In this game, players compete to build the most extensive train network across the United States, requiring strategic decision-making to win. However, with multiple strategies available, one key question arises: which approach is the most effective?

Our research aimed to develop an accurate model of Ticket to Ride, incorporating all essential gameplay elements—train pieces, cards, routes, turns, and scoring. To test different strategies, we programmed artificial agents with distinct decision-making processes and ran hundreds of simulations where they competed against one another. By analyzing performance metrics, we identified the most successful, reliable, and complementary strategies, offering valuable insights into optimal gameplay.

For a deeper understanding of our methodology, findings, and conclusions, refer to the video presentation and academic paper shown above. Additional resources are available below, with further context into this Modeling and Simulation course available in the Coursework section of this page.

Computer Vision Change Detection

Summary

This project explores cutting-edge techniques in image change detection, leveraging partial differential equations to refine computer vision capabilities. By analyzing two images captured at different times, the algorithm intelligently filters out incidental variations while emphasizing significant transformations within the scene. With an advanced weighting system, it enhances the later image to reveal persistent patterns, making this approach a powerful tool for applications in surveillance, medical imaging, and environmental monitoring.

The research began with establishing the necessary mathematical foundations, formulating a multivariate, second-order Euler-Lagrange equation, and computing the relevant partial derivatives for the analyzed energy functional. These mathematical principles were then implemented in MATLAB, allowing rigorous testing of the algorithm on example imagery.

For a deeper dive into the methodology, calculations, experimentation, and conclusions, refer to the brief slide preview above. A full version of the project presentation is available below, and further description of the topics covered within the PDEs for Image Processing and Computer Vision course are present within the Coursework section of this page.

See More From This Project
Full Slide Deck Presentation

Coursework

Adaptive Filtering

With a comprehensive foundation in stochastic processes, this course covered a variety of advanced adaptive filters including: LMS, nLMS, RLS, steepest descent Wiener filters, Kalman filters, and Voltera filters. These powerful tools were applied to solve real-world challenges such as enhancing GPS signal accuracy and mitigating intersymbol interference (ISI) caused by channel distortions, showcasing their practical impact across diverse applications.

Fourier Techniques and Signal Analysis

Widely utilized across engineering disciplines, foundational Fourier techniques were covered in this course including: Fourier transforms and series theorems, moments of inertia, band-related filters, and the critical relationship between sampling, Nyquist criteria, and signal reconstruction. Leveraging these tools is essential for advanced signal and image processing applications such as communications systems and synthetic aperture radar (SAR) imagery.

Advanced Computer Architecture

Designed to bridge the abstract software components and physical hardware features, this course explored characteristics of modern CPUs such as: five-stage pipelines, branch prediction, data & control hazards caused by out-of-order execution, benchmarking, memory access time, and multi-core systems with routing strategies. Studying and simulating CPU elements through C++ programming has been instrumental in gaining critical insights into hardware behavior which forms the backbone of developing software that efficiently harnesses the full potential of the underlying computer architecture.

Mathematics of Machine Learning

Essential for developing effective artificial intelligence and machine learning algorithms, the topics covered in this course provide a mathematical foundation consisting of advanced linear algebra, bases vectors, eigenvalues and vectors, weighted least-squares, regularization, RKHS, and singular value decomposition. Mastery of these topics enables informed design decisions when creating neural networks such as GANs and RCNNs using tools like Keras and TensorFlow in Python.

Advanced Digital Signal Processing

Digital signal processing techniques for both unidimensional and multidimensional signals, including imagery, offer a robust suite of tools such as: linear time-invariant (LTI) systems, up and down sampling, Riesz bases, Z-transforms, JPEG compression, and Tikhonov reconstruction. These methodologies are widely utilized for frequency analysis, filtering, signal and image compression, as well as reconstruction, showcasing their versatility and critical applications across diverse technological fields.

Modeling and Simulation

Building on the foundational concepts introduced in the University of Florida's undergraduate modeling and simulation course, this advanced course deepens understanding and provides new focus points such as physics & mathematics-based modeling, agent behavior, iterative simulation techniques, modeling randomness, and data visualization strategies. A strong background in modeling and simulation enables the ability to anticipate system behavior, refine performance, and implement data-driven decisions, significantly enhancing problem-solving capabilities and efficiency in designing complex applications.

Advanced Programming Techniques

Integrating theoretical instruction with a series of hands-on, project-based labs, this course provided an in-depth understanding of industry-leading techniques. Key topics included high-performance computing with CUDA, graphical programming using OpenGL and SFML, and advanced client-server relationships. Through the programming of C++ on Windows, I further honed my skills in approximation methods and efficient code design, enabling the development of high-compute physics simulations for grid environments and visually engaging windowed programs and games with dynamic graphical displays.

PDEs for Image Processing and Computer Vision

Leveraging Partial Differential Equations (PDEs) for image processing and computer vision provides powerful mathematical tools to analyze, enhance, and reconstruct visual data using tools such as: the Laplace equation, Von Neumann analysis, energy functionals, gradient descent, image denoising, calculus of variations, and active contours. Mastering these concepts enables more efficient image segmentation, noise reduction, and feature extraction, supporting advanced applications in computer vision, 3D stereo reconstruction, and image registration.

Computer Network Security

Grounded in real-world systems, this course offered a comprehensive overview of essential digital security elements including: TCP/IP layers, cryptography, DNS/DNSSEC, TLS/HTTPS, web attack strategies like DoS and their defenses, email security, and the dynamics of malware and botnets. Mastery of these concepts equips me with the skills to develop resilient applications, protect sensitive data, and proactively address vulnerabilities in today's interconnected and increasingly complex digital landscape.

Probabilistic Graphical Models in Machine Learning

Focused on representing and reasoning about uncertainty in complex systems, this course covered key topics in advanced machine learning including: Bayesian networks, Markov models, and Gaussian processes, exact and approximate inference methods, sampling methods, parameter and structure learning, and nonparametric Bayes method. Mastering these concepts enhances decision-making, predictive modeling, and structured learning in applications such as natural language processing, computer vision, and robotics.