Spring 2026 Colloquium

Blair Hoyt is a Principal at Harness IP and a registered patent attorney with more than 20 years of experience advising technology companies on intellectual property strategy.

With a background in electrical engineering, he has drafted and prosecuted patents across diverse technologies including AI/ML, electric vehicles, communications systems, semiconductors, imaging technologies, and agricultural innovations. Blair’s practice spans portfolio development, freedom‑to‑operate analyses, due‑diligence reviews, M&A support, litigation assistance, and strategic counseling. He chairs his firm’s USPTO Practice Committee and serves on both the Loss Prevention and Professional Development committees. Blair is admitted to practice in Virginia, the U.S. Court of Appeals for the Federal Circuit, the Supreme Court of the United States, and the U.S. Patent and Trademark Office. 

He earned his Electrical Engineering degree from Virginia Tech in 2003 and his law degree from the Antonin Scalia Law School at George Mason University in 2008.

Patent law was rarely discussed in engineering classrooms, yet it relies heavily on the same analytical rigor and technical understanding as engineering. In this talk, Blair will describe his transition from a Virginia Tech ECE graduate to a career in patent law. Blair will explain how core engineering skills—critical thinking, systems analysis, and technical communication—translate directly into patent practice, and how those skills can support long‑term professional growth outside conventional engineering roles. He will also share personal lessons on career development and mentorship, offering practical guidance for those considering alternative paths while still leveraging their engineering education.

Yue Cheng is an associate professor of Data Science and Computer Science at the University of Virginia (UVA). He received a Ph.D. in Computer Science from Virginia Tech in 2017. His research interests include AI systems, systems for AI, serverless computing, and data and storage systems. His group has built a number of techniques to improve the sustainability, efficiency, and scalability of cloud and AI platforms.

Some of his works have led to large-scale deployments and adoptions in public clouds and power the AI applications used by millions every day. He is a recipient of several highly competitive awards and honors including an Amazon Research Award (2020), an NSF CAREER Award (2021), a Meta Research Award (2022), the 2022 IEEE CS TCHPC Early Career Researchers Award for Excellence in HPC, and a Samsung GRO Award (2023).

Large-scale model hubs (e.g., Hugging Face, Kaggle) and numerous private repositories collectively host millions of pretrained and fine-tuned models, primarily large language models (LLMs). As the de facto infrastructure for AI model development and sharing, these platforms support a vast ecosystem of downstream applications across research and industry. However, their storage footprint has grown explosively; Hugging Face alone hosted over 70 PB of model artifacts in late 2025 and continues to expand exponentially, posing mounting sustainability challenges.

In this talk, I will present a new perspective on sustainable, large-scale modern AI model storage that rethinks system design from the ground up. I will first show how fine-tuned models within a model family exhibit high latent similarity, enabling storage systems to move beyond generic compression toward model-aware data reduction. I will then talk about why model-level assumptions fundamentally break down and why storage systems must be redesigned around a tensor-centric abstraction that minimizes redundancy at the tensor level. Finally, I will share a vision for a tensor-centric AI ecosystem built around the tensor-centric storage infrastructure.

Justin is a digital leader driving Full Visibility’s expansion into the Department of Defense/War and Intelligence Community mission areas. Justin started his career as a software developer before advancing to a senior executive in a large technology firm before joining Full Visibility. He has demonstrated a deep expertise in technical program leadership, business development, and business operations while driving innovation and delivery success in complex and dynamic operational environments.

Outside of Full Visibility, Justin loves spending time with his wife and three children, coaching flag football, steaming crabs, and running. He is a National Down Syndrome Society Athlete Ambassador and spends time raising awareness of Down syndrome.

Justin holds a Bachelor of Science in Computer Science from Virginia Tech and a Master of Business Administration from the Smith School of Business at the University of Maryland.

In this talk, Justin Knapp shares his journey from a budding developer to a senior technology leader, highlighting the personal and professional decisions that shaped his career.

Drawing from his experiences in industry, life beyond work, and his time at Virginia Tech, he reflects on the lessons learned about building strong relationships, committing to meaningful hard work, and leading with respect, integrity, and empathy. Through stories, practical insights, and actionable advice, this talk offers graduate students in electrical and computer engineering and computer science a perspective on how to define success on their own terms and lead with purpose in an ever-changing industry.

Jason Bower is a 2002 Virginia Tech Computer Engineering graduate with 23 years of experience leading innovative technology teams across Microsoft and Amazon. Currently serving as Director of AWS Field Support for North America, he leads an organization dedicated to helping customers successfully build, transform, and run their businesses on the cloud.

Jason began his career at Microsoft as a Program Manager on the .NET Framework SDK and Windows Mobile, building and leading international teams across the US, Japan, and China focused on input technologies including text prediction, handwriting recognition, and speech recognition. His early work on predictive text and small language models for embedded systems earned four patents and shares parallels with the large language models used by generative AI systems today.

At Amazon, Jason drove the launch of new Kindle devices across large cross-functional organizations spanning hardware, embedded software, and web services. Jason transitioned to AWS in 2012 where he helped build a support organization that works directly with AWS's largest enterprise customers to design, deploy, and operate some of the world’s largest and most critical workloads.

As a father, technologist, golfer, coach, mentor, and leader, Jason brings a unique perspective on how mission-driven work, cultural alignment, and strategic timing create meaningful impact in technology careers.

What does it take to go from a Virginia Tech ECE graduate to leading teams that support some of the world's largest cloud infrastructures? This talk shares hard-won lessons from over two decades of building transformative technology at Microsoft and Amazon—including the time we nearly broke Amazon.com during the Kindle Fire launch.

You'll hear real stories about the smartphone wars at Microsoft, where we built ultra-small language models that predicted text with just 200KB of storage (sound familiar to today's LLMs?), and why we lost to iPhone and Android. You’ll hear about the early days of the cloud, how scale breaks everything, and how the lessons we learn enable larger innovation. Whether you're wondering if your first job will be your dream job (spoiler: it probably won't be, and that's okay), trying to decide between competing offers, or curious about what it's really like to work on products used by millions, this session will give you practical frameworks for navigating a career in tech. We'll explore how mission-driven vision matters more than you think, and why the AI transformation we're living through right now mirrors the smartphone revolution I experienced firsthand.

This isn't a polished corporate presentation—it's a candid conversation about real projects, real failures, and real decisions that shaped a career in tech. Bring your questions about AWS, cloud architecture, career paths, or what it's really like to work at companies where one wrong configuration can take down half the internet.

Madhu Athreya is an engineer at Google, leading the development of ISPs for Pixel cameras, TPU architectures for on-device Gemini silicon, and Tensor security. He also heads advanced R&D in Video ML and Gen AI. A Virginia Tech alumnus, his career includes roles at Intel, HP Labs, and Samsung Research. With a background spanning ML, computer architecture, and signal processing, he has navigated the industry's evolution from the "AI Winter" to the current generative AI explosion.

When I was a student at Virginia Tech, the "AI Winter" was setting in. My father advised me to focus on evergreen fields like signal processing, and to strengthen math & engineering foundations. While my interests led me toward then-niche technologies like Neural Networks, VLIW architectures, and Optical Computing, the underlying tenet of focusing on engineering fundamentals has defined my career and remains essential for the future.

My professional journey has taken me from VLIW processors at Intel to image processing, robotics, and neuromorphic computing at HP Labs and Samsung Research. Today, I see these "past" technologies forming the backbone of the AI revolution. From designing TPU architectures using systolic arrays to the re-emergence of optical computing for efficiency, the tools have evolved, but the fundamentals remain the same.

In this talk, we will explore,

• The Silicon Reality: How AI is reshaping computer architecture, briefly covering bottlenecks and how emerging technologies such as in-memory compute are solving these.
• The Future Applications: How these same engineering fundamentals apply in the future —from practical robotics (drawing on my experience at HP Labs) to future humanoid assistants, and even into Life Sciences with breakthroughs like AlphaFold.
• The Career Strategy: Why uncertainty is the only certainty, and how a deep focus on engineering first principles is the best way to "future-proof" your career.

This session is a call to optimism: you do not need to predict the future to be ready for it; you simply need to master the fundamentals that will build it.

Stephen Moore is a Founder and the Chief Technology Officer of AlphaSix Corporation, where he leads work in large-scale data systems, cybersecurity, and advanced analytics in support of U.S. federal government missions. Over a career spanning nearly three decades, he has built and led engineering teams delivering production systems across domains including intrusion detection, malicious code reverse engineering, and enterprise-scale analytics.

A Virginia Tech alumnus (B.S., Computer Science, 1995), Mr. Moore began his career in information security research and operational defense, working with organizations such as the National Security Agency and MITRE before co-founding Crucial Security, which grew to over 100 engineers and was later acquired by Harris Corporation. He also holds a master’s degree in Computer Science from Johns Hopkins University. Moore is the inventor on a U.S. patent for a novel anomaly detection technique and has architected multiple large-scale distributed data platforms using modern open-source and commercial technologies.

A proud Hokie, Moore remains actively engaged with the Virginia Tech community. He regularly attends Virginia Tech sporting events, is married to a fellow Hokie, and recently celebrated his daughter’s graduation from Virginia Tech.

As AI rapidly accelerates software development, many engineering students are questioning what this means for their careers. This talk argues that while AI makes generating code easier, it makes strong engineering discipline more important than ever.

AI is powerful but unreliable — it forgets context, ignores constraints, and simulates work without executing it. Drawing on real-world experience building large-scale systems and engineering teams, the talk shows how disciplined engineering practices turn AI into a force multiplier rather than a risk, and why these skills remain a long-term career advantage for ECE and CS graduates.

Garrett Campbell is a Senior Software Engineer at Microsoft. As a software lead on the Visual C++ IDE team, he has contributed to the C++ developer experience in Visual Studio and in VS Code. In the time that Garrett has been with Microsoft, he’s made key contributions supporting Embedded developers in VS Code such as the Serial Monitor and supporting C++ developers who use the CMake or Makefile build system.

Recently, Garrett has played a key role in integrating AI into C++ developer tools as AI continues to impact developer workflows. Specifically, he has worked to develop tools to improve build performance, integrating existing build performance tracing capabilities with AI.

Prior to joining Microsoft, Garrett graduated from Virginia Tech in Computer Engineering with a focus on Software Systems.

In this talk, I will focus on the importance of always learning and the ability to adapt to ever-shifting priorities and technologies. Throughout my career, I’ve worked on a myriad of different build systems and in a variety of languages and frameworks. Over time, it’s shown time and time again that the ability to learn on the fly is crucial and I will share my experience as I’ve navigated this in my career.

Now, in this new world where AI is creating an even faster moving landscape, this is more important than ever. Especially in the developer tools space, where many companies and products are making rapid changes to take advantage of new technologies to help developers work faster and more effectively. This talk will discuss how developer tools are changing and how AI can be leveraged to improve developer workflows.