5 years after PhD defense.

Today is the 5th anniversary of my PhD defense. I was wondering if I should write some kind of reflections post, I started and deleted it over the weekend. However, while on call with a project collaborator this morning, we ended up chatting about our personal journeys while waiting for others to join - that prompted me to write again. Usual disclaimer: I am writing it for myself, and not intending to be too critical about myself.

In the past 5 years, it seems like I explored all sorts of post-PhD careers: 4 months as a post-doc (in Germany), 2.5 years as a tenure track assistant professor (in US), 1 year as a data scientist in industry, and finally, the past 1.5 years as a researcher in a government organization (in Canada). What did I learn? What did I achieve? What did I give back? I feel like these past 5 years can be put as: “walking in and out of comfort zone (mostly out)”.

I moved to Germany (from India) in 2011, to pursue my PhD. The first two years were spent more or less in the same milieu as my masters. Bunch of people with similar backgrounds, working on Natural Language Processing topics. The second half of my PhD working under the supervision of Detmar Meurers was spent in exploring new possibilities. I ended up collaborating with researchers from backgrounds as diverse as cognitive psychology and educational psychology, and even successfully publishing with them and writing small grant proposals. There were times I felt like my future career will go no where with this kind of research, but eventually, I felt confident about starting a tenure track position in US, in an even exotic (for me) setting - applied linguists.

I think this tenure track life in a department where I felt like an outlier was truly far away from the original comfort zone. Yet, the inter-disciplinary background prepared me to setup both successful and unsuccessful collaborations with researchers from a wide range of backgrounds - journalism to veteriniary science, linguistics to computer science. Teaching was as much tiring as it was fun, and even here, I explored teaching a wide range of student audience e.g, teaching NLP to graduate students coming from English teaching background, undergrads from a range of liberal arts disciplines, and graduate students in computer science. I left too early, but I felt this has been the best time of my professional life - I had sufficient independence to pursue what I want, and yet, everything was substantially challenging (from getting a good computing server and finding good students to getting stuff published) enough to not feel bored.

Scene moved to Canada by mid-2018 - I was quite grumpy for leaving what I thought was my ikigai, but I realized there can be many ikigais :-). After some years of calling myself inter-disciplinary, and several years in academia, I started my life in Canada as a data scientist in a software engineering setup, building small NLP focused teams. This is still outside my general skillset, so I was quite skeptical whether I will fit in. But thanks to good managers, I felt like I made a easy transition from academic research to industry. I actually reached a conclusion that academics can be good data scientists.

Eventually, I moved into my current job, which is technically my original professional milieu. Another externally funded inter-disciplinary project with a team of economists started by this time - I am enjoying playing that most ignorant person in the project about economics, who also happens to be the only NLPer. Then, I got into co-authoring a book, which I consider as something outside my comfort zone… not the writing part or book part, but the fact that I am writing something for industry professionals, without much experience.

Several years back, I found this quote in a machine learning journal:
”..for a true interdisciplinary collaboration, both sides need to understand each other’s specialized terminology and together develop the definition of success for the project. We ourselves must be willing to acquire at least apprentice-level expertise in the domain at hand to develop the data and knowledge discovery process necessary for achieving success.”
(from: “Machine Learning for Science and Society” by Cynthia Rudin and Kiri Wagstaff, Machine learning, 95(1), 2014)

Several times in the past years, I thought of this, and I sort of consider this a cardinal rule for inter-disciplinary research in my personal work ethic :-). So, if I reflect on these 5 years (and the 2 years before that), this is what I tried to learn, and achieve. This is also what I want to give back to the community eventually - some meaningful inter-disciplinary work that is usable for either research or practice.

Written on July 27, 2020