The Coq mailing list

[Samuel Gruetter <gruetter AT mit.edu>]

Hi Carl,

I'm only a grad student, so I don't know how researchers admit students, but when I was looking for a master thesis internship in academia, I also cold-emailed a professor at a top university, and ended up getting an internship offer. My application contained the usual material, but also a link to a small but non-trivial pull request to one of the professor's group's Coq projects on GitHub, and judging from what I heard later during a social hour, that did make a difference.

I also heard anecdotes of students who read the papers of the researchers they wanted to work with very thoroughly and found some technical mistakes, and emailed the author to ask why the paper said this surprising thing which they rather would expect to be that other thing, got a reply saying this was indeed a mistake, followed up by asking if there were PhD positions available, and eventually got admitted to the PhD program.

Good luck!

Sam

On Fri, 2020-04-17 at 18:23 +0000, Carl Patenaude-Poulin wrote:

Hi all,

I am a young software developer with a hobbyist's interest in formal methods,

including and especially Coq. I feel like I've taken what I could from working

in industry as a software engineer, and I am extremely interested in

transitioning to a career in research. I am humbly asking for pointers on

entering the field of PL, Coq and formal methods. I am also trying to adjust

my judgment with respect to what's realistic and what isn't for me.

My academic pedigree is underwhelming; I have an undergraduate degree with

middling grades (GPA 3.24/4.00), and do not have research experience. I am

confident I can do better after maturing for a few years in industry, but I am

not sure how to best make that case.

The first thing I've tried was cold-emailing a few researchers who I would be

enchanted to work for/study under. I've consciously aimed too high by writing

to those whose work I found the most interesting; not coincidentally, lots of

other people would like to work with them, and I'd imagine they have their

pick without taking a chance on cold emails from unknown undergraduates. I did

not receive strong returns, and halted this approach after I saw this bit on

Philip Wadler's page (http://homepages.inf.ed.ac.uk/wadler/phd.html) which

suggested that I had not been demonstrating sufficient effort in the emails I

sent:

>When you apply, pay particular attention to the research proposal. The

>purpose of this is not to fix an area of study, but to assess your ability to

>select a suitable research problem, to justify why it is interesting, and to

>express yourself clearly.

On to some questions:

- The blurb from Wadler's web page implies that cold emails should include a

  research proposal with a non-trivial amount of detail. Should I pick a

  problem and start working on it _before_ I apply for a PhD program?

- What approaches might I use for finding potential advisors that would be

  open to taking a chance on an unproved student? Surely it would be uncouth

  to email-blast the >100 authors of papers I've found interesting in the past

  few years? I don't want to make a bad name for myself when I've not even

  started working in the field.

- What should I include in such an email? I could put my CV and/or my

  undergrad transcript, would that be helpful? Should I attempt to signal that

  I've meaningfully engaged with the recipient's research (which I have,

  otherwise I wouldn't be writing to them)?

- One of my professors from undergraduate once offered to recommend me to a

  researcher they know, someone I would be enchanted to study under - this is

  probably a best case scenario, right? I should probably start by pursuing

  that?

All of my gratitude for your time and attention,

Carl Patenaude-Poulin

B Eng Software (McGill University)