Dr Jason Ku
✏️ I am Dr Jason Ku, a Senior Lecturer here at NUS, and I enjoy teaching students! Before coming to NUS, I was at MIT for 15 years (2005-2020), first as an undergraduate, masters, and PhD student in Mechanical Engineering, then as a postdoc and lecturer in the department of Electrical Engineering and Computer Science. A fun fact about me is that I am an origami designer, which informs my research designing structures that fold and transform. I currently serve as Chairman of the Board for OrigamiUSA, an origami organization in America. Here is my website (unfortunately still hosted by MIT web servers): https://jasonku.mit.edu/
❓ What was the most memorable experience you had while in university?
✏️ Of course academics was the major focus of my time as an MIT student. But my most memorable experiences were with the numerous student groups I was involved with: multiple singing groups (MIT Logarythms, MIT Chamber Music Society, MIT Chamber/Concert Choirs), a residential living group (Theta Chi Fraternity), and the MIT origami club (OrigaMIT). Participating in these groups helped me grow as a person, from a very shy first-year student into a group leader and someone comfortable performing on stage in concerts nearly once a week. Those experiences taught me valuable lessons in public speaking, diplomatic communication, and how to work effectively with others.
❓ Share one interesting thing that you have been working on recently!
✏️ This summer, as part of my research, I developed a piece of software called "flat-folder" that can exhaustively compute all possible valid folded states of a flat-foldable origami crease pattern, using a new algorithm that, for many inputs, is provably faster than previous algorithms. One might think that just knowing the location and direction of folds on a paper (e.g., a crease pattern) would be enough to determine how to fold an origami model; but actually, even deciding whether a given crease pattern can fold flat without intersection is NP-hard  (read as "widely accepted to be intractable to solve using modern computers"). But more importantly, "flat-folder" produces pretty images of foldings for real origami models, which I post daily to an instagram account. Here are links to the project's github repository and instagram account:
 Akitaya, H.A., Cheung, K.C., Demaine, E.D, Ku, J.S., Horiyama, T., Hull, T., Tachi, T., Uehara, R. (2016). Box Pleating is Hard. In: Akiyama, J., Ito, H., Sakai, T., Uno, Y. (eds) Discrete and Computational Geometry and Graphs. JCDCGG 2015. Lecture Notes in Computer Science(), vol 9943. Springer, Cham. https://doi.org/10.1007/978-3-319-48532-4_15
❓ What is one piece of advice you would like to offer to ME students?
✏️ A degree in Mechanical Engineering is globally recognized because it provides an excellent perspective of the world, together with a standard set of tools that can be used to approach and answer many problems. But every discipline has its own perspective and tools, and it can be useful to gain many perspectives. These days, Computer Science is very popular because almost every discipline (in both industry and academia) is using computers to advance their fields.
✏️ At NUS, computing skills are primarily taught by the School of Computing, so incoming students think they need to major in Computer Science to learn those skills. But computing as a field is different from computing as a skill. As an analogy, ME students take classes in mathematics, but that doesn't mean they should all become mathematicians. The reality is that most students want to learn computing skills to apply them to applications in other fields; for ME students, those fields might be Robotics or Machine Learning.
✏️ So my advice to ME students would be to supplement your ME degree requirements with a single additional module: specifically CS2040C, an algorithms course that teaches you how to reason about computer programs and to argue that they are correct and efficient. (CS2040C is a version of CS2040 targeted to students outside of SoC.) It provides a different and valuable perspective that will make you a more effective Mechanical Engineer, with the added benefit of preparing you for programming interviews in case that is relevant to you. Full disclosure, I am biased, since for 3 years before coming to NUS, I taught 6.006, the MIT equivalent of CS2040. You can find recordings of the last term I taught 6.006 on OCW:
Leave a Reply.