Graduate Q&A: Kasimir Gabert

The College of Computing confers degrees to deserving graduates every semester. During the week leading up to spring Commencement, one graduate reflected on his time at Georgia Tech.

Kasimir Gabert graduated with a Ph.D. from the School of Computational Science and Engineering (CSE) where he was advised by Professor Ümit Çatalyürek. In a Q&A discussion, Gabert described his best memories and offered advice on “getting out” of Tech.

What were your research interests and notable projects while at Georgia Tech?

We are living in a world with an increasing amount of data. From diverse areas such as social activities, biological processes, and computer traffic, much of this data contains internal relationships which are well represented as graphs. I have been fascinated with the problem of trying to understand the internal structure of these large, sparse graphs, especially as they change over time. I have been involved with two notable projects. First, discovering a way of keeping track of nuclei, which are locally dense and important regions of graphs, as the graphs are changing. The second has been to develop a large, distributed system that can compute analytics on changing graphs and elastically scale itself as the rate of change and size of the input graph varies.

What is your favorite memory from Georgia Tech?

Every year at the annual supercomputing conference, Georgia Tech holds a booth and typically has several excellent papers. I have great memories of staffing the Georgia Tech booth, attending engaging talks, participating in various supercomputing events, and making many new connections with people. It is hard to justify numerous, year-round in-person computer science conferences with their associated carbon footprints and travel demands; having one annual, domain-wide conference may be a sustainable approach that avoids many of the drawbacks.

What advice would you give to other students who are just getting started on their Ph.D. here?

Getting a Ph.D. is not a small undertaking and, differing from your prior degrees, it will change in shape significantly over the course of the degree. At many points it will feel indefinitely far away; you may get discouraged and frustrated with rejection and a lack of progress. My advice would be to always keep moving forward in whatever way you can. Do not dwell on setbacks, and instead always try to improve and find the most exciting, rewarding path forward. If possible, tune your memory to preserve details on every paper you have read, and, at the same time, forget every rejection you have received.

What is next in your career? How did Georgia Tech help you get there?

I will work as a staff member at Sandia National Laboratories, addressing data science problems across a variety of domains. My experience at Georgia Tech has been crucial for me being able to address such problems effectively. My advisor and committee have helped guide my research so that it can apply to real-world problems, both by addressing the scale of the data algorithmically and through developing well tested and usable implementations.