Grad School Series: Oxford University
Arijit Patra graduated from IIT Kharagpur in 2016 with a dual degree in Mechanical Engineering. He is currently pursuing his PhD at the University of Oxford, where he studies machine learning and biomechanical models that can predict Congenital Heart Diseases from fetal echocardiography videos. He went to Oxford as a Rhodes Scholar (one of the 89 elected worldwide for the class of 2016).
When did you decide to pursue a PhD? What motivated that decision?
My situation was quite unorthodox as compared to the vast majority of undergrads seeking avenues for higher study. I wasn’t particularly decided or interested in an MS or a PhD. I did, however, have a history of research projects and an FT (in lieu of the Honda YES Award), but none of that was motivated by an overarching goal of a PhD.
I applied for the Rhodes scholarship in my fifth year, primarily because of its reputation and my love for proper nouns. Once I had it, I had to decide on a course of action. One thing led to the other and I started a PhD program at Oxford!
How did you decide on colleges to apply to? For someone unsure about what they want to do, is MS a better option?
Based on my experience of observing people’s trajectories, I would suggest a matrix of choices. This is something most people already do, in terms of safe, moderate and ambitious university/program choices. However, I’d also suggest choosing programs beyond the acceptance difficulty and this largely depends on your intended career path. For instance, if one wishes to do an MS with an eventual goal of converting to a PhD (say in an area like computer vision), it is not always essential to secure admission to a top 10 institute in the broader field of computer science. There could be individual labs that could be competitive in particular sub-fields and a Master’s thesis with such a group could open many doors for a top-ranked PhD program later on.
For someone unsure, an MS does sound less demanding, but the fees and lack of support funding may be a deterrent.
Can you comment on the relative importance of CGPA, Statement of Purpose (SoP) and Letters of Recommendation?
Never ignore the power of a good CGPA! That said, your CGPA isn’t everything as applications are almost always evaluated holistically. So, if you are lacking in some criteria like CGPA, you can always compensate for that by research projects/publications and relevant internships.
A good SoP always helps the admission committee contextualize your application more optimally and personal motivations behind the intended courses should find expression at this stage. This is not something your resume can cover.
How did you go about choosing your advisor? Are there opportunities for funding once on campus?
The funding aspect vastly differs from place to place and department to department. While I didn’t do it the conventional way given the unusual trajectory that I had, I would suggest choosing supervisors after perusing their research, talking to their students/alumni and studying the career paths of their alumni.
Any parting words for current undergrads on how to approach the entire apping process?
I didn’t really do much of the ‘apping’ process, so I’m probably not the best person to answer this question. However, from what I have observed over the years, it helps to seek the guidance of immediate seniors in your intended fields and seek their feedback at different points — early on, when performing exploratory analysis, in mid-stages, when drafting SoPs and later on — when processing app decisions.
All said and done, here are pointers based on what I think I should have done during my time at KGP:
1. Participate in as many research projects as possible — particularly in those labs which have a strong publication/patenting culture.
2. Look to take as many Math/CS courses as possible — common knowledge in these fields are increasingly making headway in most other domains (case in point : the recent surge of deep learning in turbulent flow modelling)
3. If a certain lab is not working out (maybe you’re putting in too much effort without sufficient hope for tangible results), pivot quickly — time is limited
4. Do a foreign internship early on. I know there could be better avenues for serious research in Indian universities as well, but the experiential value of surviving in a foreign land early in your undergrad gives you a quiet confidence that becomes indispensable later on.
5. Get some industry-relevant skills (data analytics etc) and possibly do small projects for startups or larger companies. This teaches accountability. In a similar vein, I think a year or two of professional experience in the industry is beneficial for further studies (PhDs in particular) because it is really important to get used to real Mondays after 4/5 years of vacation!
