Exchange Diaries  

Arnab Maiti | Computer Science and Engineering

University of Washington, United States of America

While you were in college, what was your motivation to go for higher studies and in
which year did you get that motivation? How did you manage your academics with your
preparation?

This usually happens with the kind of exposure you get. I have been in LBS hall
since 1st year, and there were a lot of seniors who were motivated towards PhD. I
luckily came in contact with those kinds of seniors and seeing to them I got
motivated. I actually got motivated in the first year not straight away for PhD but
at least try out research as it suits me or not. I also used to attend the seminars
that IR Cell conducted every year where they called the students who got PhD
offers. After the seminars I came in contact with the seniors especially from my
department and I used to have constant chats with the seniors and this way I got
interested in research.

Regarding the part of managing the academics, I did an intelligent thing. As in
third year we have electives. I chose electives which were aligned with my
research area which helped in academics as, whatever I used to read for research
it was more than enough for the electives. Due to the pandemic there was no
academic activity, so I got a lot of free time to do research work. I was lucky that I
got many holidays after 3rd and 4th year due to corona. Usually IIT do not give
such a long break where I get enough time to do my research work as, during the
semester it is difficult to do.

What are the factors that you considered for selecting the universities you are going to
apply for?

One factor everyone should consider is reputation of the institution. The second
factor is the area of interest. In PhD the area is quite important. It's not just the
brand of the institution, area is also important as the university must have the set
of professors which you are targeting. Brand is also important to some extent. It
does not mean that brand should override area but, after PhD if you want to end
up with R & D, or you want to end up being a professor, so the University plays a
very important role. I considered just the United States for my PhD reason being
that computer Science is pretty much based in the United states Internationally in
forum.

Did you apply to multiple universities? If yes can you please elaborate on the differences
in the selection process of the universities?
In universities in which you were not selected can you please share what you felt you lacked,
and what could you have done better?

Yes, I applied for 9 Universities out of which I got 4 PhD offers

The starting phase is similar to every university, like you have to fill all your
details in the application portal. You need to write TOEFL, it is mandatory and
GRE is not mandatory. You need to apply your CV, which needs to be research
oriented. You need to find 3 letter writers, so 3 professors who are willing to
endorse you, usually that comes from the research project with whom you did.
Now, in some institutes the graduate committee, which is the pool of professors
and some senior PG students who decides whether to select the candidate or
not. Other institutes are decentralized and they leave it to professors and they
ask the professor to take an interview and give the feedback, they then go back
and report to the graduate committee.

I applied to MIT, Stanford, CMU, Berkeley, University of Washington, UIUC,
Georgia Tech, Columbia and University of Pennsylvania. So, MIT and Berkely
usually, very rarely Indians get into those institutions, also they need to know your
letter writers, so my letter writers were Indian professors so I think they even
didn’t even know who my letter writers were, so that was the main reason I felt
that I was rejected. Stanford took my interview 2 times, but there was an area
mismatch. The thing on which I want to work on exactly there wasn't any
professor, they also offered me to work in other areas, but I wanted to stick to my
area, so it was a rejection. Columbia and Pennsylvania I was not sure why I was
rejected, but sometimes, they get to know that this guy has got into a top
University, so they move their offers to others. Now the remaining four
universities have offered me a PhD.

How are you planning to manage your finances? Have you applied for any scholarships?
What were their procedures?

PhD is funded completely. I’ll be getting 3200$ per month. Your stay will take
1000$ to 2500$ and food will take 500$, so you can manage within 200$.

No, I haven’t applied for scholarships and most of them are not for international
students.

Is CGPA a factor while considering higher studies at foreign universities?

As long as you have a decent 9.5 CGPA, you need not have a 9.9 CGPA to get a
PhD offer. Even 9+ CGPA works. More important is the fact that the area in which
you are applying should have a good grade, preferably EX. It's better to have good
grades in the courses in which you are applying for PhD.

Have you taken up any research project at your bachelor/master's level under any
professor here at IIT KGP? How did you manage your college studies and worked for
your research?

Yes, I have taken. I have done 3 complete projects with prof. Palash Jain. In our
department, we have a project seminar, then a B.tech project and then M.tech
project.

Academics, I didn't compromise anyway. I sacrificed my social life, didn’t get
involved in any societies and also not even in any research group at IIT
Kharagpur. I was completely focused on academics and projects. I took courses
which were aligned with my academics. During the semester I was more focused
on academics and, in terms of research work, I was thinking about my projects.
As thinking was not time consuming, I used to reserve time like one hour in a day
reading or thinking, and the time consuming part of writing papers I did it in the
summer breaks. My research work did not involve any experimental work. It was
more statistics oriented and algorithm oriented. That’s the way I could manage
academics and research.

What were the other difficulties you faced on taking this path and how to overcome
them?

You’ll face multiple challenges. The first challenge is self doubt which I
overcame when my mentor assured me that it is common. Then, another
challenge is peer pressure like your peers getting placed in top companies, or
having a good internship. I overcame it by comparing myself with my previous
version of me and this way I focused on my work. Another thing is that you get
less sleep, as you have to work long hours. As you sleep late at night and get up
early for classes. So If there are classes I used to sleep early and if there are no
classes next day I used to sleep late. Another challenge was covid pandemic as
in campus you have friends you can rely on, which provide emotional support, but
at home I used to listen to music or watch anime or movies to keep myself
relaxed. Keeping yourself relaxed is important, it shouldn’t be that you keep on
working continuously. I tried to work at night and during the day I used to do fun
activities and physical activity and used to sleep for 6 to 7 hours.

How did IIT KGP benefit you in your journey?

The best thing about IIT KGP is actually the students. As I got interested in
research the reason is because I saw my seniors doing the same kind of stuff. At
IIT KGP has a research culture so I got to expose it. IIT KGP takes students to
research projects, so these were the things where I got benefitted. Seniors end up
going to foreign universities so they maintain the good reputation of IIT KGP.
Many foreign universities like CMU have many IIT KGP students. It has a positive
impact on you, so this also benefits you. IIT KGP doesn’t force students to work
at IIT especially for the 4th year summer, as it is not allowed by other IIT’s, which
helped me to explore new areas and work with another professor at IISC.

What are the opportunities after the Ph.D. course?

There are two broad opportunities in computer science: you can become a
professor at a top university or you can enter some research industry. Few people
join quant companies where they appreciate PhD students.

If you did PhD from CMU then, you can become professor in any top University in
the United States or you can go to Europe or you can become professor in India
as well. For research Industries you have Microsoft Research, Facebook
Research, Google Research. There are other kind of research industry and there
are ML based companies in the United States. ML based companies have lots of
subdivisions on which they work like deep learning, reinforcement learning,
computer vision, natural language processing, pure theoretical computer science,
computer systems so, all these areas are there they actively hire students from
these top universities so that they get to work in a research group. Usually 50%
you have to do research work and 50% you work for the industry in the
companies.