Welcome to Engineering Innovations, the official podcast of Purdue University's Elmore Family School of Electrical and Computer Engineering. I'm your host, Kristen Malavenda. I'm the Communications Director for the school. In each episode of Engineering Innovations, I'll sit down with faculty from Purdue ECE to talk about their path to becoming an engineer, the focus of their research, the future of the field, and more. And joining us today is Millan Kulkarni.
He's a professor, but also the interim head of Purdue ECE. Thanks for joining us today, Millan. Really glad to be here. So let's start with that first thing I mentioned, the path to becoming an engineer. What's the earliest you remember doing things that would later lead you down the path to being an engineer? So I have, I think what from the outside looks like an incredibly stereotypical story of becoming an engineer. Both my parents are in technical fields.
In fact, my dad was a mechanical engineer as an undergrad and industrial engineering and operations research for his PhD and actually so did my mom. So engineering kind of runs in the family. It's in the blood, so to speak. And so from an early age, of course, when instead of getting toys that, you know, balls and bats and stuff like that, we were not an athletic family. Instead, we would get, you know, erector sets and things and building toys of various sorts.
I remember getting this really cool little plastic gear system called Capsula where I could build little boats that had propellers that could like run through the bathtub. So from a very early age, I was always about building things and taking things apart. The particular kind of engineer that I am is a computer engineer. And I'm from that generation where it was actually not that common when you were, you know, when we were young to have computers at home.
But now, of course, everybody has computers and we got our first computer when I was seven years old. And was that unusual for the time? It was right around when that was starting to happen. If your parents were in a technical field, maybe it was more likely for you to have a computer, but most people maybe didn't have them. They maybe had them at school at best. So I was seven years old and my dad, because he's an engineer, taught me how to program basically from the beginning.
This was a very old computer. There wasn't that much you could do with it, but you could write computer programs. And so I learned to program when I was quite young. And so like I said, stereotypical story. Learn to program when I was young was always into computers, always into taking them apart, rebuilding them, building my own computers. So when I got to undergrad, it was a natural choice to become a computer engineer. And it kind of progressed from there.
So when you finished your undergrad, decided to go undergrad school, what turned you towards academics as opposed to industry? Yeah. So this is a, again, this is one of those stories where if you maybe look from the outside, it actually seems obvious that I would go into academia. There's this kind of common statistic that goes around that a huge fraction of people who are in academia have parents who are in academia. And that's true for me. My dad's a university professor.
So you might look at that and say, well, obviously, Miland was going to become a university professor. But that actually wasn't obvious. Even when I was an undergrad, it wasn't clear that I was going to go to graduate school. Certainly that was the wish that my parents had for me as people that had gone to graduate school. But I was an undergrad during the height of the dot com bubble. So why would you go to grad school when you could join a dot com startup and making millions of dollars? Right.
So maybe luckily for me, but maybe not the dot com bubble burst right before I graduated. So I graduated in 2002. The dot com bubble had just burst a year or two ago. And so grad school kind of seemed like an obvious option. So I went to graduate school. I was getting my PhD in computer science. And even then it wasn't actually clear that I wanted to go into academia.
I sort of enjoyed what I was doing, but it wasn't, I don't know that I loved doing research at least in the first few years of being in graduate school. And so I was thinking, well, I'll get my degree and then go off to industry. Go to Microsoft, go to Intel. At the time there was this up and coming startup company that a lot of people were going to called Google. Maybe I wouldn't be doing this podcast if I had done that. I'd be on an island somewhere instead.
But I, by about year four of graduate school, I realized that research was actually something that I really enjoyed doing and teaching was something that I really enjoyed doing. And so at that point academia became the obvious destination. And speaking of research, like where did you start with your research and where are you today and what research you're doing? Yeah. So this is, I'll back up to grad school again.
When I was getting into graduate school, I did what I would actually tell students now to never do, which is that I wrote my statement of purpose saying, you know what, I actually, I like a lot of different fields in computer science. And so I don't really know what I want to do. I don't know if I want to do computer architecture, which is basically, you know, hardware, right? Or artificial intelligence. Like those were my two choices, kind of on the two opposite ends of computer science.
And that's how I wrote my statement of purpose. And I'm honestly just lucky that somebody said, despite that terrible statement of purpose, that I could, you know, join grad school. But when I was in graduate school, I took a class in compilers from a professor, Keisha Fengali. And I loved the class. It was the first time that I had seen a class where deep mathematical concepts could be applied to real world engineering problems. And it fascinated me, right?
This idea that I could take these mathematical concepts that seem completely unrelated to the problem of compiling. And for those of you that don't know, compiling is basically the problem of translating a computer program that you might write into what the computer understands so that it can actually execute. Got you. But the idea that you could do this, that you could take these deep mathematical ideas and use them to drive how you translate a computer program into machine code, blew my mind.
And it didn't help that that that professor, Keisha, was was a fantastic teacher. So that first year I asked him whether I could maybe do research in compilers, which is something I had never considered before. And that's kind of how I got into compiler research. When I was in graduate school, I did a lot of work on automatic parallelization, the idea of finding opportunities to run different parts of a program simultaneously so that everything runs faster.
When I got to Purdue, I started working on something a little bit different, but kind of still broadly in the same space, which was how can I take the kinds of computer programs that you might see in fields like computer graphics or in fields like data mining, that compilers historically have not done a good job of targeting and make them run faster by doing various kinds of compiler transformations, by basically restructuring the program so
that it runs better on the computer hardware that you might have.
And the reason that I really like this field of research, which is honestly with slight variations what I continue to do today, the reason that I like this field of research is it requires deep understandings of applications, of understanding the kinds of algorithms that people are running, what it be like, what somebody that's writing a high performance ray tracing engine really cares about, while also having a deep understanding of computer hardware.
So what do you have to do to make a program run fast on a modern computer, on a modern GPU? So you have to understand both of these sides and kind of harkening back to graduate school where it was, how do I take these deep mathematical insights and make them work for compilers? This is about how do I take these deep algorithmic insights that a lot of people have and translate them into something that a computer understands well. Gotcha. And I just love that challenge.
And I know from just interacting with you that you're a professor who really enjoys teaching undergrads, how do you use your research and your just experience to teach your undergrad classes? Yeah. So this is, I love teaching. I tell people that it's because of teaching that I would never leave faculty.
You could go to industry research labs and do research, you could go to national labs and do research, but there's nowhere else that you could really get the experience of sitting in front of hundreds of students or standing in front of hundreds of students and teaching them something they didn't know before. And nothing can replace that. One of the, so I teach, when I teach undergrads, I teach lower level courses. I teach, you know, intro to computer programming.
We call it, well, advanced C programming is any other course. And discrete math, which is kind of the foundational math course that computer engineers need to have. And one of the interesting challenges in both of those courses is that it can be really hard for students to understand how these very fundamental concepts that we're trying to teach them translate into what they actually need to know in the real world. You might look at a math problem and say, this is math.
What does this have to do with what I'm going to do when I graduate and go work at a Google or Apple or a Microsoft? You might look at a really basic C programming assignment and say, well, this is incredibly basic. It has nothing to do with, you know, the Python code I'm going to write if I'm doing some deep machine learning work when I graduate. And so for me, a big challenge is how do you make those fundamentals interesting and applicable for students?
And I've actually found my own research to be really effective there because it's very easy for me to draw direct lines between some of the fundamental computer programming stuff that I might teach in C programming to, hey, these techniques actually show up in these real world applications and understanding how these techniques work is how you can make progress.
Using a discrete map, hey, these mathematical techniques that seem really abstracted and really almost obscure, they actually really show up in my research and other people's research. These things are at the cutting edge of what people care about in the field. And so being able to make to draw those connections is really helpful. It shows the students that what they're learning isn't just abstract. It isn't just something we're making them do.
It really does undergird so much of what they're going to do in the future. So you're not too far removed from your studies, but I'm sure that what undergrads go through these days in engineering is different than what you did. What are those differences? So some parts of electrical computer engineering are fundamental.
When I teach discrete math, yes, some of the applications that I might talk about today are different than the applications I might have talked about 20 years ago or my teachers back then might have talked about. When I teach C programming, same deal. Some of the applications and some of the stuff at the frontiers looks different, but a lot of the fundamentals are still the same. Circuits are circuits. A C program is a C program. Discrete math is discrete math.
Where the differences really start to show up is when you start thinking about how those things are applied. Just as a simple example, when I was an undergrad, I took a course in artificial intelligence. Those are senior level electives. And a thing that literally never came up in this entire semester of learning about artificial intelligence was a neural network. We didn't talk about it at all.
We focused on what in computer science we call symbolic artificial intelligence, because that's what people were thinking about. Now the idea that you could take a course in artificial intelligence and just never talk about neural networks or honestly not spend a huge fraction of your time on neural networks, it was my mind. Yeah, right. The field moves fast at the frontiers. The foundations stay the same and the frontiers shift quickly.
And that actually makes for a really exciting educational experience because you're always thinking about how to tie those foundations to those frontiers. So honestly that's one of the big changes is not so much the content of your sophomore level courses. It's the content of those junior and senior level courses that has really shifted quickly.
But honestly one of the things that I think really cool about our field is that you're able to do these junior and senior level courses that do start talking about things at the frontiers of the field and those foundations still work. Right? Right, they're still what you need to know. So how you mentioned that the field is just constantly changing. How do we as Purdue ECE keep up with that and make sure we're teaching the students what they need to know today?
Yeah, so this is a great point and it's something that we think about a lot. So there's a few things that I would say. So one, I honestly believe that this is the reason that research universities work as well as they do. It means that when we're getting folks in to a university like Purdue where we're teaching thousands of students, the people that are teaching the thousands of students are people who are at the forefront of their fields.
They know what the modern world needs in computer engineering or in power systems or in automatic controls. So they are defining that modern world and then they turn around and they get to talk to juniors and seniors about how the modern world works. So I think it's really important we have these phenomenal researchers at a university like Purdue in a school like ours that can then turn around and teach these students. I think that's one of the ways that we keep those connections.
One of the other things that I think is actually really phenomenal about our department that helps build these connections is the students take on a lot of the work themselves. The kinds of projects that people do, whether that's through class and things like senior design or VIP or EPICS, or whether it's on your own through the numerous student groups that come out of ECE, it blows my mind.
The level of complexity, the level of rigor, the amount of cutting edge technology that people are deploying in these projects, it's mind blowing. It's amazing. I know every year we have the Spark competition twice a year, which is our senior design kind of showcase and I'm constantly amazed. I say you built that in a semester and yeah, they're incredible. Yeah, I wish I had a senior student's capacity to operate on very little sleep.
Yeah, well I asked the team that won last year's Spark competition. I said you built this in a semester and they said yeah, I'm like how? And they said honestly, we don't remember. Sleep deprivation is how. But no, I look at what I did 20 years ago in undergrad and what our students are doing now and there is no comparison. So beyond kind of the students taking on a lot, what makes Purdue ECE special or stand out from other programs?
We are constantly rated in the top 10 and so what puts us there? Yeah, I mean this is one of the things that we're proud about. We are a very large program and despite being an incredibly large program, we are excellent. We're constantly rated in the top 10 both for grad and undergraduate education.
I think a big part of it is that despite being a top research program, we have a lot of faculty that really care about education and really care about making sure that students are getting the right material and having that material taught effectively. There's, you know, I could go down the list and think about people like Professor Sundaram or Professor Brinton or Professor Chan, you know, across our department.
There are these faculty that are both excellent researchers and excellent educators and I think that really can shine through and can get students excited about this. I think another aspect that really helps us here is we have a phenomenal instructional staff that works closely with faculty and is constantly talking to faculty and amongst themselves about how we can provide a great educational experience at the scale that we are.
Right, I go and talk to people at other departments and I tell them, yeah, our circuits, you know, our fundamentals one class, what other schools might call circuits one, we have a thousand students a semester. Wow. And it blows their mind. It's inconceivable that you could do, that you could be the scale that we are and do as well as we do. But I think that's a testament to our faculty, our staff and again our students for really taking on that challenge.
So your passion for teaching is very apparent. Then several months ago you decided, hey, maybe I'll be the head of the school too. So what, why did you take on the position of interim head? Yeah, I ask myself that sometimes. Why do I do this? I will say the thing that I miss the most about being interim head is not being in the classroom and not being able to teach. I did keep one foot in teaching. I still advise an epic section.
So I still, you know, on a weekly basis get to interact with students and talk to students about what they're working on and their passion. Part of it is that we are a very large school and there's a lot of moving pieces and the way that we can succeed is by keeping all those moving pieces working together smoothly.
And so it's important to have somebody, I think, you know, one of the things that I really enjoy about being interim head is the opportunity to see all of these different threats that are constantly, you know, flowing in our department and understanding how they all tie together and sort of making sure that everything stays tied together and thinking about how the research mission feeds into the teaching and the teaching feeds into the research and all of the different pieces.
So it was a phenomenal challenge. And I also stepped into some very big shoes. Our previous head did a great job of elevating our school. And so it was a little daunting to take it on. But I'm glad I tried and I'm glad I did. It's been a lot of fun and it's really great. I get to, as interim head, brag about what everybody else is doing. And there's no greater sense of accomplishment than being able to talk about your fantastic colleagues and students.
So we're coming off a pretty several years where we had a lot of growth, both in terms of facilities and in the classroom. What do you see as the next challenges and things that Purdue EC will be tackling? Yeah. So there's a few things. So, yeah, here at West Lafayette over the last 10 years or so, we've seen a tremendous amount of growth in the department, way more faculty, way more students. Our physical plant, our buildings are starting to keep up.
We have this great new project going on right now where we're completely revamping our undergraduate instructional labs. I'm really excited about what that's going to look like next year. And I'm really jealous of the experience that our students are going to have starting in the fall. We show that project to some of our alumni. This is the first floor of the BHW building. We show that project to some of our alumni and they say, well, this isn't fair. They're going to get windows in the labs.
Right. We didn't have windows. And then some of them turn around and say, well, you can only do a great job in an ECE lab if you don't have windows. Right. You feel like you're in a cave. Exactly. So we'll see how that part plays out. But I'm really jealous of some of the facilities there and excited about how that's going to take us to the next level. I think one of the big challenges that's coming up in the next year is where our growth is going to be now is not here in West Lafayette.
Right. We're launching, you know, Purdue is launching this ambitious new mission down in Indianapolis, right? A new Purdue campus that is still us. Right. It's still our department, our unit, but also teaching folks down in Indianapolis. And so it's a new territory. Right? It's not, I wouldn't say it's green field because we're right in the middle of a big city. Right.
It's a new opportunity for us and a new phase of growth for us where we, I think we will still be growing in terms of faculty, in terms of students, in terms of facilities, but in a very different way that we have in the past. Not in our traditional footprint, not in places that we've all become familiar with over the 120 plus years that the department's been around, but in a brand new space. And that's really exciting. It's a little bit scary. It's ranking and a little bit scary.
But I think the opportunities there are really great. Yeah. It's another thing that could set us apart from other programs as well. So Purdue Compute is a big thing. Yeah. Big. I don't know that we can even, we can probably do a whole podcast just about Purdue Compute. But it feels to me like ECE is going to play a major role in this effort that the university is spearheading. What is our role in Purdue Compute?
Yeah. So Purdue Compute is this kind of umbrella term that the university is placing over a bunch of initiatives that are all kind of tied together by their importance or by their foundation in computing. And so this involves both growth in terms of core computing work, the kind of stuff that we do in computer engineering and also other parts of the department and ties with computer science, which is of course also where core computing work happens. But it also involves semiconductors.
And all of the investments of the university is making in semiconductors, understanding that there's a vital national need for excellence in that space, both in terms of research, but also in terms of workforce development, putting out students.
And if the country is going to come and say we need tens of thousands of new semiconductor engineers, well, what better place to do it than a place like Purdue where we have fantastic semiconductor researchers in classes and we just so happen to educate thousands of engineers. It also includes artificial intelligence. This is revolutionizing the world.
And there's opportunities, I think especially at Purdue, to think about how artificial intelligence is going to revolutionize the physical world, the things that we touch and grow and move, not just computing. So this is the iPod, the Institute for Physical Artificial Intelligence that Purdue has, Purdue ECE has a huge role. It also includes quantum. So we're talking about things, Purdue gets to talk about doing things at very large scales, but we also do things at extremely small quantum scale.
And Purdue has a lot of excellence there. So this is Purdue Compute, this big umbrella thing that spans maybe these four things, including AI, semiconductors, quantum. When I look at that, what I find fascinating is that there is not a one of those four things that our school is not at the center of. If you want to do computing, you should come talk to us. If you want to do AI, especially AI as it applies to real world things, you should come talk to us.
If you want to do semiconductors, well, you should definitely come talk to us. And if you want to do quantum, you should come talk to us. This is, we are at the heart of what the university is trying to do across all of these initiatives. And it's really exciting. I don't think that these initiatives can be successful without us and without a successful ECE. And having said that, ECE is obviously a huge part of it, but there are other departments on campus.
I think it's kind of a microcosm of how research and teaching works now. It's very interdisciplinary. We're not in our silos anymore. How much of that is important to what ECE does? So before I was interim head, I was the associate head for teaching in the department for several years. And as part of that, I got to interview basically every faculty candidate that came through. And we've been in a growth phase. We've interviewed probably over 100 faculty candidates in that time.
And one of the questions they would always ask me is, what is one of the things that you really value about ECE? The thing that I would consistently tell them, and you could go talk to some of the assistant professors that we've hired over those last years, and they'll confirm this story. One of the things that I would consistently tell them is just how much our department values collaboration.
Just how much we think it is important that our faculty work together to tackle the emerging challenges of the world, whether that's working with an ECE or working between ECE and physics or ECE in mechanical engineering or ECE in computer science. We have a very collaborative department. It's one of the things that I've loved about this school in the 15 years that I've been here. And I agree. This is the way that we're going to make progress on some of these grand challenges of our time.
There are very few things out there where you could say, oh, I can solve this fundamental problem while staying in my silo. Yeah. Yeah. It's not as much fun either. That's right. It's working with other people. So, I mean, it's not like you don't have a lot of responsibilities between the little teaching you do, the interim head. When you do have free time, what kind of things do you like to do? Well, I would have answered this question very differently 10 years ago.
What I like to do in my free time is, you know, I love cooking. And so I'm kind of the primary cook for our family. I read a lot of science fiction and fantasy, but also other things. I play the piano and the mandolin in my spare time. Cool. So, that's a lot of fun. I love watching movies and going to see movies. Eight years ago, the story really changed, which is that now I've got an eight-year-old and an almost six-year-old. He's turning six in a little bit over a week.
And they take up a lot of the time. Do you see engineering seeds in them? Are you doing to them what your parents did with teaching the coding? Yeah. I'm doing, I'm engaging in a little bit of brainwashing. My wife is a social scientist. She's a psychologist. So, she's not the biggest fan of my attempts to kind of guide them in an engineering direction. But you know, my daughter, who's eight years old, above her bed is a little collage of prominent women scientists and engineers.
Oh, I love that. Right. So, Rosalind Franklin and Grace Hopper and Hedy Lamar. And so there's a little collage of that. It started early. When she was two years old, she had to do a little project in her daycare about what she wants to be when she grows up. She said she wants to be an engineer. I definitely saved that and put that somewhere safe. So, I definitely see those seeds. I don't want to force them down anyone path. If they want to be engineers, then I would love that.
Yeah. And I do love that they are really excited about how the world works and science and math and things like that. But they have so many talents and there's so many different amazing things that they could do. So, I'm not going to try to put them into a box. Gotcha. Well, I think that's all I have. Thanks so much. It's been a great time talking to you today. Thanks for joining us. Yeah, this was a lot of fun. Great.
So, that's it for this episode of Engineering Innovations, the official podcast of Purdue University's Elmore Family School of Electrical and Computer Engineering. If you liked the show, please rate it and subscribe and make some comments. It really helps us. Your feedback helps us craft how the next episodes are going to be. And taking you to the next episode, we'll have another one next month. Until then, thanks for joining us.