Things I learned from my 7 interviews- Omkar Shetye (ICTS, Banglore) | PAE blog

I was asked to write about my interview experiences but I didn't want to make 7 different posts so I have decided to write one essay summarizing all those experiences. I have been to interviews at the following institutes:
TIFR-Mumbai(I-PhD and Ph.D.), ICTS(I-PhD and Ph.D.), IISc(I-PhD), IISER-Pune(I-PhD), IUCAA/NCRA(Ph.D. preselection)
Amongst these, I cleared TIFR Mumbai (IPhD in 2016 and Ph.D. in 2019), ICTS(Ph.D. in 2019), and IUCAA/NCRA(December 2015).

So, instead of telling you stories of those experiences, I have decided to write about what I learned from those interviews.

1. Soft skills play a role albeit a really small one. They aren't as important as the case is in a corporate job interview, but the way you speak and dress always, even if subconscious, has some effect on the interviewer's mind. This does not mean that you have to dress up in all formals. A well-ironed shirt with jeans works well for me. Although I did clear the IUCAA interview wearing a t-shirt, so it's not going to make or break your entry into the institute. The same goes for hair/beards. I have cleared interviews with neat hair, big ugly beards, and also completely bald. So, I don't think they really have a preference here. Try not to sound too braggadocious when you speak because the people in front of you have the skills of a professional roaster and will bring you down very quickly(they have very creative insults).

2. Everyone who doesn't think that the world revolves around them will find their interview experience slightly disappointing irrespective of whether you clear it or not. "I should've done this better", "I could've solved this in XXXX manner rather than what I did", etc are common thoughts. Just stop thinking and wait for the results. After my IUCAA interview, I was so disconcerted that I ended up screwing up my TIFR entrance which was just a day later. The very next day I got an acceptance mail from IUCAA. So, don't let interviews bring you down.

3. Give as many exams as possible and appear for every interview you can even if you don't want to actually join the institute. Experience will teach you how to control your nerves and think on the spot. Way too many people I know think "I am not prepared enough this year, I will give it next year." GIVE THE DAMN EXAM. Even if you get a negative score you will have gained something positive from the experience at the end (what doesn't kill you---something----).

4. Every interview begins with the question: "What subject/topic do you like?" This is going to be the most important question you answer and will decide almost half the questions that the panel asks. I can't press hard enough how important it is to not screw up questions on whatever topic you put up here. Do not say any topic you don't know enough about. I generally say Quantum Mechanics here. Sometimes they will ask you to give one more topic of interest too, so keep a couple of subjects handy. Since you guys come from engineering backgrounds, I would mention that I have seen Engineering people generally go with Thermodynamics and Computation as their first choice of subject. (Some go with QM/StatMech and impress the hell out of the panel too, so you can risk it if you are confident)

5. Do Maths properly. Almost every interview will consist of some maths question even if a simple one and if you are, say, unable to plot log|x| then your chances of getting through takes a really big dive.

6. It is good to have an idea about what your recommendation says about your skills because you don't wanna be surprised by a "Your referee thinks you are good at Stat mech so let's start there". Legit happened to me once. I suck at Stat Mech.

7. By the end of the interview, the panelists start running out of questions to ask and you might face a couple of weird ones. It is quite okay if you don't answer all of them.

Questions I faced:
I always said I was interested in QM so most of my questions consisted of solving square well potentials, harmonic oscillator solutions, perturbation and in case of Ph.D., finding partition functions, F-D, and B-E statistics, etc. Questions start as a textbook and then the panel thinks of making it more and more complicated as you go on solving them. These are some really good questions I was asked:
For I-PhD:
1. (IUCAA)How to find the temperature of a star? <follow up> What will you do if you have a single data point on the energy spectrum of the star?
2. (ICTS)Find the low-temperature behavior for the specific heat of a two-level system.
3. (TIFR)Why is the resistance of an electric bulb calculated from the power rating different from its resistance measured by a DMM when it is off? <follow ups> How will you find the temperature of the filament? If you use the Black body radiation method for finding temperature, what is the expected answer? Why is this wrong? Why is this discrepancy present? And finally, what gives rise to resistance in solids?
4. (IISc) If you go on decreasing the depth of a finite square well potential, at what point does it stop having a ground state? (this was the follow up after asking me to derive energy spectrum for finite square well)

For Ph.D.:
1. (TIFR) Does the energy of a muon affect its lifetime?<follow up> The lifetime of muons affects the spread of their energy spectrum. If you increase the average energy, will you detect muons with a sharper energy spectrum?
2. (TIFR) Explain the interference pattern for Young's double-slit experiment far away from the central maxima.
I haven't mentioned other questions because they were pretty standard ones.    

About the author:

Mr. Omkar Shetye is about to begin his Ph.D. in ICTS, Bangalore, one of the best research institute in the country for Physics.

Also, do visit all other useful content of our blog Physics after Engineering.

Below is a page with the link for our WhatsApp group with 500+ members all of whom are engineers who actively discuss and help each other out in entering the physics stream after engineering.

https://physicsafterengineering.blogspot.com/p/blog-page.html