Kevin Hung, a true New Yorker at heart, was born and raised in New York City, and still lives there now. Kevin received his bachelor’s degree in computer science and economics at New York University. After that, he worked at a financial services company for about a decade, first as a systems administrator and then as a software infrastructure engineer.
Interestingly, Kevin originally enrolled in Georgia Tech's on-campus M.S. in Computer Science program in order to pivot from fixing things to building them. Georgia Tech was his top choice due to the program's abundance of AI-related courses, along with the acclaim and research output of the university's College of Computing. He was aiming for a career related to artificial intelligence, as the field was starting to boom and he had always been interested in the idea of computers being able to do something without being explicitly told how to do it. In fact, Kevin hopes to work on explainable AI, where the AI system can provide an explanation for its answer that a human would accept. This has not been the focus of most recent AI research and development since neural networks, the technique du jour, have poor explainability. There is an ongoing shift in this attitude though, as more and more people have started relying on AI systems for decision-making, making the desire for explanations more important. Moreover, to Kevin, computer science is the study of how to make computers an effective tool for our needs. Computers are typically a means to an end, not an end unto themselves. The need can be "better tracking of business expenses", "simulating of scientific phenomena to prove a theory", or even "entertaining". All of these are grounded in the field of computer science. “What I love most about working with computers is that it scratches my itch for tinkering in a frictionless manner. I like to compare it with Legos: you get building blocks that you can arrange however you like. If you don’t like the result, you can simply snap off the unwanted pieces and make something else. With computers, you can replace parts of programs or extend them however you like just by adding or removing lines of code.”
"What I love most about working with computers is that it scratches my itch for tinkering in a frictionless manner. I like to compare it with Legos: you get building blocks that you can arrange however you like. If you don’t like the result, you can simply snap off the unwanted pieces and make something else."
During Kevin’s first semester at Georgia Tech, the COVID-19 pandemic hit, and his classes went online. The sudden shift in format was challenging for some classes. He also found that he could not keep up with a full-time graduate-level workload, since he wanted time for self-study and research into topics not covered in class. The option of OMSCS meant that he could take similar classes that were already prepared in a remote-friendly format. In addition, he was also able to drop down to a more sustainable workload without worrying about the cost/benefit trade-off of effectively being a part-time student. Furthermore, what he likes most about OMSCS is interacting with the teaching staff and with other students in the program. “This helps to foster better understanding of the material. The obvious way this happens is by asking questions that I can then get answers to. However, thinking about how to answer a question another student has posed can also help solidify my own understanding of the material.” Kevin’s favorite class has been CS 6601: Artificial Intelligence. “While challenging, the projects and the exams really complemented the lecture material by making you work through the details of different algorithms. Oftentimes, it’s easy to fall into the trap of thinking that you understand the material because it makes sense at a high level. However, when you actually dive into the details, you often learn that you don’t understand it as well as you thought you did. That class definitely helped me avoid that trap, or perhaps it’s better to say it made me step into and out of that trap multiple times.”
Besides computer science, Kevin is also interested in both physics and philosophy. Both attempt to explain the world from different points of view. Physics tries to explain the world in terms of its mechanisms, while philosophy tries to explain the world in terms of how humans view and understand it. Ideally, after graduation, Kevin would find a job where he can work on AI. Besides development of machine learning models themselves, there is an ecosystem of jobs surrounding the usage of such models. This includes ML Ops, which involves deploying and maintaining machine learning models, and data engineering, which involves creating systems to collect and prepare data for the models to use. Kevin is in his last semester, so the current job market will impact how his plans unfold. “There have been many layoffs in Big Tech recently, so there could be fewer jobs and more competition than usual. An alternate plan is to find a job unrelated to AI, then transition to one that is related within the same company.”
"Thinking about how to answer a question another student has posed can also help solidify my own understanding of the material."
An interesting fact about Kevin is that he does not have a driver’s license, as he considers driving a car as more of a liability than an asset in New York City. However, he also argues that the convenience of a car elsewhere means he is eagerly awaiting the widespread adoption of self-driving cars. Also, if Kevin could collaborate with anyone, past or present, he says it would be with the noted computer scientist and naval officer Grace Hopper. “Her idea that humans should use words rather than symbols to write computer programs led to the development of the programming language COBOL, which is still in use today. This concept of user-friendliness led to the creation of the field of human-computer interaction (HCI), where computers are not considered in isolation but with respect to humans that use them. I believe this to be a core part of the acceptance of much technology, and it would be amazing to see what her input would have been in HCI-related subjects such as explainable AI.”
As for Kevin’s favorite hobby, it is reading. He is partial to science fiction, but he also reads current events, non-fiction, fantasy, and classical literature. Kevin does not limit himself to the English language either; he also read Japanese light novels such as Kino’s Journey in the original Japanese!