Incoming PhD in Computational Materials Science at target school in UK

[computationalmaterialscientist]

Hey so I am about to start my PhD in Computational Materials Science (i.e. physics) and I wanted to join the community to learn about what I can do to get a really great head start for quant roles after my 4 years. I am going to a strongly targeted UK university (one of Oxbridge, UCL, ICL, LSE, Warwick) and my project involves inverse problems in machine learned interatomic potentials. I will be writing new packages to add to existing atomic simulation environments in Python too.

Anyway, as far as I know my project will be quite well aligned with quant researcher/developer roles, but I want to build up a portfolio of projects so that I will be prepared for jumping straight into a quant role as soon as I finish my PhD. So, is there any advice for projects to start with? Thanks!

 

[Andy Nguyen]

Welcome to QuantNet. You have a great start by joining us.
4 years is a good stretch of time to get familiar with different roles, firms, skills required.
I would start with the list of reading materials (many are free) here. This gets you the basic lingos and traditional roles.
Strongly consider getting a strong C++ experience. Many interviews are asking C++ data structure.
C++ and Python will open way more doors than only one or the other.

 

[computationalmaterialscientist]

Welcome to QuantNet. You have a great start by joining us.
4 years is a good stretch of time to get familiar with different roles, firms, skills required.
I would start with the list of reading materials (many are free) here. This gets you the basic lingos and traditional roles.
Strongly consider getting a strong C++ experience. Many interviews are asking C++ data structure.
C++ and Python will open way more doors than only one or the other.

Thanks for the advice. I will check out the reading materials and look at C++ learning resources.

 

[Daniel Duffy]

Computational Materials Science (i.e. physics)

What is the focus? mathematical or more experimental?

In general, physics is not (necessarily) maths. Not a bad thing. Just saying.

 

[computationalmaterialscientist]

Computational Materials Science (i.e. physics)

What is the focus? mathematical or more experimental?

In general, physics is not (necessarily) maths. Not a bad thing. Just saying.

So my part is more mathematical and computational. I won't be doing any experiments, but I will be using mathematical models in the code that I'll be writing.

 

[Daniel Duffy]

This book might give some relevant info

https://onlinelibrary.wiley.com/doi/book/10.1002/9781119719731

And C++ is kind of vital.
Are you using Fortran?

 

[computationalmaterialscientist]

Thanks, I will check out the book.

In my PhD I will have the opportunity to do a module using Fortran for high performance computing which I was planning on taking. However, all the code that I will be writing for my project will be Python.

I am more than happy to learn and build up a C++ portfolio on the side though.

 

[Daniel Duffy]

Thanks, I will check out the book.

In my PhD I will have the opportunity to do a module using Fortran for high performance computing which I was planning on taking. However, all the code that I will be writing for my project will be Python.

I am more than happy to learn and build up a C++ portfolio on the side though.

Fortran is the best for HPC (I started with Fortran 66..)
It is a niche, however.

Before building a C++ portfolio, first learn C++ properly, no half measures.

It might come as a shock that C++ has no native Matrix structure.