Dr. Anton Dahbura is a professor in the Computer Science Department at Hopkins. I currently work with him on Delineo, a disease modelling project in collaboration with the World Health Organization and Ruvos.
This interview has been edited for brevity and clarity.
Can you tell us a little bit about yourself?
My name is Tony Dahbura. I am on the faculty of the Computer Science Department. I also co-direct the Johns Hopkins Institute for Assured Autonomy, and I direct the Information Security Institute at Hopkins.
I read that you were also an undergraduate student at Hopkins. What initially drew you to Hopkins, and what about it here made you stay even as a professor?
My story is very different. I was here for undergrad and grad school. I grew up in El Salvador. I was born in the U.S., but when I was young, my family moved to El Salvador for business opportunities. In those days, there was no internet or satellite TV. The world was a much less connected place. I went to an American school, and our advising office had basically one book about colleges, and I had a Rand McNally map book. I knew I wanted to go to a school in the most selective category because why not? And I played baseball. I was crazy about baseball and was a big Baltimore Orioles fan. So, I said, what’s the most selective school that’s close to where the Baltimore Orioles play? And it turned out to be Hopkins. I applied Early Decision and was accepted.
Being an Early Decision applicant, I didn’t apply to any other schools. I knew I wanted to be here. I was very enthusiastic. I also knew I wanted to be an engineer because my dad was an engineer. When I was in high school, we would buy these kits to build a calculator or a shortwave radio, things like that, so engineering was kind of in my blood. I did my undergraduate work here. Then, I joined the BS/MS program. I joined a little late—I was a senior—but I still saw tremendous advantage in that. Then, I started working with my advisor on the side. Before I knew it, I was his PhD student. I don’t even remember how it happened. I went through the process quickly, graduated undergrad in ‘81, got my masters in ‘82, and PhD at the end of ‘83. I was fortunate—I had really good guidance from my advisor. He gave me a lot of attention, which I think is something that Hopkins provides. And I was able to solve a 15-year-old outstanding problem that a lot of people were trying to solve.
That’s interesting. What was the problem?
It had to do with distributed fault tolerance. It was kind of a liar and truth teller problem. The entities are able to query each other and determine if they’re faulty or not. The idea is if you’re a central investigator and you get all these results that this entity thinks is faulty, but it could be lying. You don’t know, maybe that one’s lying and trying to spread false information. Despite all this information, some of it’s good, some of it’s bad. How do you uniquely determine which subset of entities is telling the truth—which one’s lying, which ones are faulty, or which ones are fault-free. That was a pretty big deal back in those days.
Going off that, was there a point when you knew you wanted to be a professor?
I knew when I didn’t want to be a professor. I was drawn to industrial research and had the opportunity to work at Bell Labs right after my PhD and do basic research. It was very academic. I was given an office and told we’re not going to tell you what to do, just go find something and work on it. I spent ten wonderful years there, but I also had very strong ideas about how to manage research. I was in my 20s and Bell Labs tended to not advance younger people that quickly. Motorola came along with a new research facility at MIT in Cambridge, and they were looking for a director. I decided to jump in and spent about six years there building supercomputers and developing parallelizing compilers for large scale parallel processors. Then, in the mid 90s, my family was doing really well here in Maryland with their commercial endeavors, and I had a young family, so we decided to move back to Maryland to be closer to them. For the next 16 years, I was involved in those businesses. I kept in close touch with my PhD advisor, and in 2011 or 2012, he was planning to retire. He founded the Information Security Institute at Hopkins. I decided I wanted to get back to my technical roots and came back to Hopkins. It took 30 years. I was away for 30 years and came back, and so I rediscovered the magic of Hopkins and all the wonderful things that are going on here.
The part I really want to emphasize is that when I was a junior, my father worked as a mechanical engineer for a corporation that made equipment for sandblasting things like helicopter blades or engine blocks and cars, and they were having a big problem because their hoses were getting clogged up. It was very serious because if a hose got clogged, the engine blocks wouldn’t be peened properly, and the operators would detect it until it was too late, so it would damage what was being worked on. They didn’t have a way of detecting that. My dad asked me to go to one of my professors and see if they would like to work on the problem. I went to one of my professors, Jan Minkowski, and explained it to him. I’ll never forget how he looked at me quizzically and said, “Why don’t you solve it?” I said, “Me?” He said, “Yeah, I’ll give you a lab. I’ll give you a little budget. Go solve the problem for them.” So, I worked very diligently, and as we always do in research, had ups and downs, trial and error, and finally came up with a way to not only detect if the hose was clogged, but actually what the exact flow was. And it took a little bit of time. The company didn’t believe it at first because it was showing some weird phenomena that was happening inside of the hose. They had to validate that, and then they were convinced. I actually started building these little boxes and selling them back to the company so they could measure the flow in their hoses. That got me hooked on research as an undergraduate, and so one of the things that I wanted to do when I came back to Hopkins and what I worked out with the dean was really being able to work with undergraduates on research ideas and give students a lot of room to explore and to have success in research. The dean thought it was a good idea, so I currently have almost two dozen research projects. The topics vary from sports analytics research and robotics to the modeling and simulation of the spread of disease, and cybersecurity. Some of it’s whimsical, some of it’s very practical. I have 70 or 80 undergraduates involved in these different efforts.
As someone who is involved in those research efforts as a student, I think it’s absolutely great. I wanted to ask about Delineo in particular. How did that journey begin?
I have an appointment in the Malone Center for Engineering in Healthcare, and in the early days of the pandemic, the leadership of that center rallied the members and said, “What can we do to improve the situation?” I still lived in Western Maryland and noticed the pandemic was impacting an area less than an hour and a half away from two major metropolitan areas differently from how it was impacting New York, Philadelphia, Baltimore, and Washington, DC. People were saying, “Oh, this pandemic is not going to reach us. It’s not gonna be that bad.” I wanted to have a way of showing how diseases can spread through specific communities, and I wasn’t really happy with the high-level modeling that traditionally is done in epidemiology. I wanted to have a way of helping people in specific communities understand how disease spreads and then come up with some strategies for mitigating it.
That makes sense. Delineo has some great partnerships like the World Health Organization (WHO), Ruvos etc. How did that come about, and how does Hopkins partner with different organizations?
We at Hopkins have a lot of great partnerships. AWS (Amazon Web Services) played a really nice role as matchmaker. They introduced me to Ruvos, and they introduced us to the opportunity at WHO. They also introduced us to NTT Data, another partner in this effort, who act as a kind of project management. They put us all together and made this happen. Ruvos, not coincidentally, uses AWS as their platform. It was a real victory for everyone.
How do you balance all your different roles at Hopkins like teaching and research projects with undergrads?
At this point in my career, I have time to dedicate to all these things. I have so much fun that, for the most part, it is not really like work. I spend a lot of time in and out of the office, working on all of these different things, and it’s still exhilarating to be able to see concepts or general ideas put into motion and become reality.
What advice do you have for current or incoming Hopkins students?
Hopkins is an amazing place. Of course, we live and breathe research, but it’s not always spoon-fed to you. You have to go out and find the opportunities. You really have to experience research when you’re here. If you go to Hopkins and you never have research, you know, it’s like going to Italy and never having pizza. You have to pay attention. You have to build relationships with your professors. Our classes are small enough that you can get to know your faculty and really get to know your academic advisor. Don’t be shy about reaching out and saying, “Hey, I really am interested in what you’re doing. How can I pitch in? How can I contribute?” Or even if you have ideas of your own, start to talk about it. The first time it may not work out, but maybe the second or third time. That’s really how it works.
My last question is can you share a favorite memory or something really impactful from your time at Hopkins?
I have a lot of favorite memories. I have so many stories, I think the relationships with my advisors have been great. I’ll tell you one quick story because that professor’s still here—Professor Kosaraju. I feel really close to him because he was my professor also, and he taught algorithms. At the end of the semester, he saw me walking down the hallway and he called me into his office. He said “Dahbura, congratulations” and it was a very begrudging congratulations. He said, “You put the optimal amount of effort into the class. You got the lowest A.” It was still an A, so it was good.
What I tell students is, for their future, be involved in research. Whether you’re going to grad school or for jobs, the one thing everybody wants to ask you about is your research project because it’s so interesting. Everybody takes calculus. They don’t want to ask you about linear algebra. Let’s say, wow, you worked on this project with this simulator and what is that all about? That really sets you apart from everyone else.
My involvement with undergraduate students is a way of paying it forward. I had a great undergraduate research experience at Hopkins, and I want as many of the students here to have that same experience I did.
