Original Link :

https://iiseralumnistories.wordpress.com/2015/07/24/abstract-math-at-mohali-a-stepping-stone-to-the-applied-world/

Nikhil Kumar graduated from IISER Mohali with a dual BS-MS in Mathematics in 2013 and joined a PhD program at TU Eindhoven (aka Eindhoven University of Technology), Netherlands the same year. This PhD opening was in part offered due to funding by the energy and petrochemical giant Shell. We talk to Nikhil about his experience.

Keshav: IISERs have a reputation of focusing on abstract maths and sciences. SHELL is an energy and petrochemical group of companies. How did you get to such an offbeat, unconventional and interesting career?

Nikhil: Shell is a big company operating in over 140 countries and invests heavily in Research & Development. Most of the key projects in Shell rely on computational results from various simulations, for which they require a pool of researchers with a sound understanding of mathematics, physics and chemistry. The course work at IISERs does appear to be abstract, but to pursue any research in computational sciences, all that stuff is needed.

K: Could you give some specific examples?

N: As an example, my PhD project is about designing more efficient offshore wind farms, which uses computational fluid dynamics (CFD). In technical terms, I am working on finite-volume methods (FVM) for solving the incompressible Navier-Stokes equation numerically. CFD in itself is a very vast field, with people using the Navier-Stokes equation, lattice Boltzmann methods (LBM) and smooth particle hydrodynamics (SPH) etc. and modelling the effects of turbulence and boundary layers in the fluid flow. In order to model all these effects, one needs to have an understanding of the underlying mathematics, like probability theory, functional analysis, differential geometry and numerical analysis, most of which is already there in IISERs’ course structure.

K: Are you then an employee of SHELL?

N: No, I am not and I have no legal obligation to work with Shell, even after the PhD. I am currently working at the Eindhoven University of Technology (TU/e). Let me explain this briefly. For the Shell-NWO programme for research in computational sciences, Shell gives funds to the National Scientific Council of the Netherlands called NWO (Nederlandse Organisatie poor Wetenschappelijk Onderzoek), which then directs the funds to its sub-organisation FOM (Foundation for Fundamental Research on Matter). I am employed by FOM and my work group, Centre for Analysis, Scientific computing and Applications (CASA) is a part of the Department of Mathematics and Computer Science, TU/e. Moreover, upon completion of my PhD, we’ll have the option to work for Shell. For this programme, Shell has invested close to 20 million euros for funding 60 PhD projects.

I would encourage maths and physics graduates to look into these options…

K: Very broadly, you are working on improving offshore wind farms, which is a renewable source of energy. How was this project decided?

N: That’s a bit difficult to answer, as the projects are decided by a panel of researchers chosen by Shell and FOM. My project is not only about offshore wind farms, most of the computational methods developed in the project can also be applied to other fluid flow problems. With that said I would like to add that although Shell is primarily involved in harnessing energy from the fossil fuels, I am sure that in the near future all energy companies will be forced towards the other side, the “greener side”.

K: That makes the picture clearer! I guess I didn’t ask you about the selection procedure. What is the competition like? Were you required to take GRE or TOEFL?

N: I wasn’t required to take GRE or TOEFL. I filled an online form and gave a telephonic interview. This was followed by interviews conducted by Shell Bangalore and Skype interviews with my tentative supervisors in the Netherlands. The selection procedure has been revised now. There were about 5000 candidates for 25 projects when I applied. I should add that Shell, as well other petroleum giants like Exxon Mobil and BP will be actively recruiting new people for the next 2-3 years. CFD is used everywhere, from airplanes to cars to ink printers and there are a lot of jobs in CFD outside India. I would encourage maths and physics graduates to look into these options.

salaries are comfortable by the Dutch standards and the job market here is good

K: Every place has an innate culture- what does it mean to be a PhD student at TU Eindhoven?

N: In the Netherlands, PhDs are employees of the universities. We pay taxes, get compulsory holidays and pension benefits just like any other employee does. Our supervisor treats us more like a colleague and the candidate has the independence as to how the project should proceed. As for the internal working, every PhD candidate gets two supervisors- a Daily Supervisor and a Promoter, who reviews progress. Most of the PhD projects at TU Eindhoven are industry funded. Our salaries are comfortable by the Dutch standards and the job market here is good. I also get to attend courses offered by Shell every 4 to 5 months. These courses are on varying subjects, like geophysics, oil extraction etc. Along with that there are many short courses offered by FOM. To top it all, Netherlands is good place to do your PhD. It’s a small beautiful country in Central Europe, with not so nice weather, where people are a bit too tall (in fact the tallest) and they speak “English” (at least most of them).

K: Apart from research, what responsibilities do you have as a PhD student?

N: 10% of our total working time goes in teaching. We need to complete 600 hours of teaching in the first 3 years and are exempt from teaching for the final years. Me and my supervisor offer an undergraduate level course on ordinary differential equations and numerical methods. Some PhDs also have to mentor master students with their thesis.

the summer projects were the best part of IISER training…

K: How did IISER Mohali prepare you for your current position?

N: In order to develop a scheme for solving a discrete system, one does not only need numerical analysis, but also an understanding of underlying analysis of the ODEs and PDEs. Apart from that it is beneficial to know the basic concepts of manifolds, functional analysis, probability and the computational aspects of the solver. The courses on discrete maths and computational methods in physics were very useful.

For me the summer projects were the best part of IISER training. The broad training at IISERs widens the knowledge base in the first two years. I don’t feel that I would have been as comfortable and confident were I in a conventional BSc-MSc Math course. E.g. people applying CFD to biological systems (like blood flow) might have problems understanding the biology of the system since they are not exposed to it, but it is relatively easier for us since we have had a basic introduction to the subject in our core years. Also, knowing a bit about subjects other than your major helps you socialize with researchers from other disciplines.

K: Apart from your training at IISER Mohali, what new skills did you have to pick up?

N: We need to do a lot of literature survey and study, as significant advances have been made in CFD over the past two decades. With the increasing dependence of industries on CFD, new techniques are discovered all the time and it is important for us to be aware of these new developments.

should not worry too much about their careers after IISER, but instead should pay more attention towards their course work

K: What message would you like to give to the IISER community?

N: I think IISERs should introduce some physics courses on Computational Fluid Dynamics. Such courses are offered in most universities in the Netherlands. There is scope for having more courses on numerical analysis. There should be at least one course for the mathematics majors which focuses on the commonly used numerical methods. It is important for industries, and thus should be important for student’s career. I would suggest adding a couple more elective courses on numerics. The existing courses could probably include mass transport equation and Burger’s equation, which is a fundamental partial differential equation from fluid mechanics. As for the current students, there are some good courses on Coursera. I would suggest the course on High Performance Computing by Dr. R.J. LeVeque. That course is just enough to build a base for a PhD in CFD.

Well that was more regarding my own work, but I feel that the students should not worry too much about their careers after IISER, but instead should pay more attention towards their course work. If you do that right, its will pay more than just sitting and rethinking about why you joined IISER.