How to apply for jobs in math

I'm getting close to the end of my postdoc and I'm applying for jobs again! So, I thought I'd write some basic guidelines that I found helpful in this process. I will update this guide as time goes by. Feel free to add your own advice in the comments.

Deadlines and reference letters

Most jobs start coming out around September for the following year. After that they keep trickling in all the way until May or even June. The first big set of application deadlines are in November, but some earlier ones are in October. Therefore, you should get your letters well before October. How many letters do you need?

  1. Around 80-90% of academic jobs need three letters. You can feel mostly relieved once you get three.
  2. Some jobs need four letters, and even specify "four or more". If you can get more than three, which is not always easy, you can pretty much apply anywhere.
  3. Industry jobs usually do not require letters for your initial application but will still require the email addresses of a couple of references.

Almost all academic jobs require at least one letter to mention of your teaching abilities. Graduate students can typically get someone in their department to write this, based on their evaluations from either courses or tutorials.

Most jobs are on so register with that site and get your recommendation letters on it. There are a huge number of jobs on this site. Typically I've seen at least a hundred postdocs and other non-permanent positions and at least another hundred permanent positions advertised.

You should adopt a triage strategy with this vast list. This year I saved all the webpages for each job into four folders from most to least desirable. The least desirable ones aren't necessarily bad ones, but they might say "all areas of math will be considered but analysis strongly encouraged". Since my research doesn't involve analysis, this one is obviously less likely than one that wants an algebraist.

In the filenames I put the deadlines first in ISO 8601 format: yy-mm-dd, so that sorting the files by name will sort the jobs by deadlines.

Jobs sorted by date in a folder.

I found this much more manageable than my previous strategy of just browsing the website randomly.

Although there are hundreds of jobs, there are hundreds more applicants! Some postdocs have more than 600 people applying for a single one. The number of people applying to each varies. For example, there were two positions at Dartmouth, and 434 people applied. Brandeis had one postdoc and 750 applicants. I wish I had more numbers to share, but suffice it to say there is still competition and you can be sure many of the applicants will be strong.

Check other websites as well

It seems like all the jobs are on MathJobs. But there are many other sites that are worth checking out as well:

Some of these might have duplicate positions on MathJobs, but many will not, especially those outside North America. It is definitely worth it to check out other countries, even though it is hard to leave your home country sometimes.

How to filter

There are many criteria that you can use to filter the huge number of jobs out there. Here are three to get you started:

  1. Location
  2. Teaching and research balance
  3. Represented research areas
  4. Faculty members you have previously interacted with

Postdoc or tenure-track?

In math it is almost impossible to go tenure-track right out of your PhD. If you can get a tenure-track position right away, chances are you don't need to be reading this guide. So right out of your PhD focus mainly on postdocs. However, if there is a really cool tenure-track job you'd love to have, try for that too just in case. After all, it doesn't take much more effort to apply to an additional position.

Sadly, if you just look at the history of current math faculty around the world, it seems that having two or more postdocs is not at all uncommon. So after your postdoc apply to everything.

Many jobs or just a few awesome ones?

Some people believe strongly that you should only apply to the positions that are really an outstanding fit. Others say apply everywhere. It is true that if some place isn't high on your list, you won't be high on theirs. Nevertheless, applying to more places won't decrease your chances of getting hired somewhere, and who knows, maybe everyone applying will be worse than you. Or maybe someone remembers you from a talk you gave. Some job is better than no job, and you can do research anywhere. Therefore I say, apply to anywhere you could reasonably work.

I aim to apply to around 80.

The cover letter

Find out everything that you like about a position, and make sure it shows in your cover letter. Committees go through hundreds of applications and anything you can do to make it stand out will help you out. This will obviously be easier for the places you really like. You have to demonstrate that you are a good fit.

Go through the prospective department's website and name the faculty whose research you are interested in. Include some specifics, including how your research might benefit the research of others in the department.

Some cover letters want additional specifics that are mentioned in the posting, like how you'll contribute to the growth of a recently created math department or how you can contribute to diversity.

The teaching statement

Here are some things to have in your teaching statement:

  1. Specific techniques you like to use in teaching, including examples
  2. Commentary on technology in the classroom: some institutions want to see this
  3. Quotations from students who filled out past course evaluations

The teaching statement should be a few pages long. Writing everything I could think of got mine to three pages.

The research statement

A research statement should be like a good general talk. The first bit of it should be understandable to almost anyone, with all the technicality near the end. It should give a solid indication of your current research along with future plans. It should be about ten pages long in the typical LaTeX document class. However, a few select institutions like Stanford have different length requirements so you should pay attention to that and have an abridged version when necessary.

The research statement should have references that include all of your papers, and it should make connections with the research of others. Bonus point if you can connect it to research of the relevant faculty members who will read it.

Industry or academia?

This is a tough question for me and one I've been debating this time around. The first time I applied for jobs, I basically ignored industry. After my first postdoctoral experience, I am considering industry much more seriously and I am applying to both industry and academia.

Before you decide, you should check out some websites of companies or recruiting agencies that don't mind hiring pure math PhDs who don't require applied expertise:

  1. Verseon
  2. Palantir
  3. Jane Street Financial
  4. Virtu Financial
  5. Bickham Montgomery

Many of these places sound like fun to work, and I've applied to some of them. CSIRO in Australia is a great place to check out for more applied jobs, though most of their jobs require slightly more applied backgrounds like statistics or modelling.

I suggest applying to some of these places and going through some of the phone or in-person interviews. Even though I really enjoy publishing research and mentoring, I have realised that it is not easy to get your research and talents recognised with a permanent position in academia, even if you work hard and publish lots of papers. Therefore, you (and I) should have some industry backup plans.

If you are reading this while you're still in your PhD, I suggest learning a programming language like Python or C or and getting involved in an open-source software project so you have some additional meat on your CV. Programming experience is favoured in many of these, and a language like Python is actually pretty easy to learn.

What to do while applying

I have found that applying for jobs takes a lot of time and research can take a hit. What I've found works for me is reading something like a long paper or a book to get some new research ideas.

Also it is a good idea to get plenty of good food, sunlight and exercise.

Leave a comment

Fields marked with * are required. LaTeX snippets may be entered by surrounding them with single dollar signs. Use double dollar signs for display equations.