Q. Please give us a brief introduction of yourself.

Hello, my name is Min-Kyu Song, and I joined the School of Electrical Engineering at Korea University in March 2025. I received my B.S., M.S., and Ph.D. degrees from Yonsei University, and then spent approximately four years as a postdoctoral researcher at MIT in the United States before joining Korea University.
My research focuses on semiconductor devices, the fundamental components used in chips for smartphones, computers, and other electronics—including transistors and memory elements. During my Ph.D. studies, I worked primarily on bio-inspired semiconductor devices that mimic the unique properties of biological materials such as proteins. Later, in my postdoctoral research, I focused on neuromorphic devices that emulate the brain, as well as 3D-stacked device architectures.
 

Q. What kind of research will you be conducting at Korea University?

While artificial intelligence software has been advancing rapidly, the hardware technologies needed to support it are evolving much more slowly—a fact evidenced by the ever-increasing cost of high-performance GPUs.
My research aims to overcome the limitations of conventional digital systems by developing analog computing devices for AI acceleration. A representative example is in-memory computing, where memory devices themselves are used to perform computation. This approach minimizes data movement and dramatically improves both energy efficiency and speed, making it a core concept in neuromorphic semiconductor design.
Additionally, I am exploring how to vertically stack these devices, similar to HBM (High Bandwidth Memory), to create ultra-dense computing platforms capable of high-performance processing in three dimensions.
 

▲ 3D-Stacked AI Hardware Developed by Professor Min-Kyu Song, Reconfigurable Like LEGO (Source: Nature Electronics) 

Q. What inspired you to pursue your field of research, and how do you view its future?

To be honest, what I found most difficult while studying electrical and electronic engineering was… coding. As embarrassing as it is to admit, one of the reasons I was drawn to semiconductors was because, among all the subfields, it seemed to involve the least coding. I vaguely wanted to try research, but as a student who disliked both math and coding, I naturally gravitated toward semiconductors without a grand motivation.
However, as I began doing research, I became captivated by the charm of academia—namely, the opportunity to develop and advance ideas that no one else in the world has explored. That fascination eventually led me to the field of neuromorphic engineering, where we propose entirely new paradigms by rethinking the fundamentals of the digital world from the ground up—materials, devices, and circuits. Ironically, I ended up doing quite a lot of both math and coding in the process.
There’s a common perception that semiconductor device research is old-fashioned or repetitive because it involves handling equipment and manually fabricating samples. But I believe the field I work in is foundational—like the bricks of a building—supporting the vast structure of the semiconductor industry. What makes it so exciting is that it allows us to draw a “big picture” that spans from device to circuit, architecture, and software. Not only is there an abundance of meaningful problems to research, but the outcomes can have a tremendous impact on industry and drive technological innovation.

Q. Do you have any advice for students?

First and foremost, I encourage you to pursue active learning. Many students at Korea University are excellent at memorizing, understanding, and solving given problems—skills honed through structured education. But what I personally found most challenging in graduate school was that research means creating your own problems and solving them.
There’s no one handing you an exam paper or a syllabus. You have to identify what to study, define the problem, figure out how to approach it, and solve it—all on your own. It’s an entirely new experience. That’s why I believe it’s essential to actively seek out opportunities during your undergraduate years to engage in this kind of thinking. I urge you to question, critique, and propose ideas in your coursework. Don't just aim to be a good student—aim to become a true engineer.

Second, until now, you've likely been evaluated by written exams, but once you graduate, you’ll find yourself being evaluated far more often through presentations. While it’s easy to think that researchers prove themselves with data and numbers, the reality is that speaking and presenting clearly matters just as much, if not more.
Some people are naturally good at presenting, and others aren’t—but I honestly believe that learning to give a good presentation through practice is far easier than getting into Korea University through academic exams. So, when opportunities to present come up, don’t shy away. I strongly encourage you to challenge yourself and develop your communication skills through every chance you get.