Recommendation letters are a really important part of my job and of your professional trajectory. I'm honored to write letters for students that I highly recommend. People that review your applications often give the letters a lot of weight (among all applicants, academic success is often a given, so letters are one of the only ways to be competitive). Letters of recommendation are a way for me to stake my reputation and goodwill on your behalf, so I want your letters to be the best that they can be, for your benefit and mine, so that my recommendations continue to be well-regarded. I've gotten to where I am in my career thanks to the people that wrote letters for me, and part of making mathematics and academia more equitable is carrying that torch forward to the students that have earned it. I don't take these facts lightly. Letters are indispensable and not "just another box to check" on your application. If you'd like me to write you a letter, here are things you need to do (and why):
- Have recently taken my class and earned an A. Academic excellence is a necessary (but not sufficient!) condition for a recommendation. "Recent" isn't a concrete thing, but if it's been a while you might need to remind me who you are, which doesn't beget a strong letter.
- Have made an impression and sought a connection with me. Letters are a chance to humanize you and discuss your soft skills, in addition to your technical abilities. A strong letter uses specific examples and anecdotes about our relationship to demonstrate why you're well-qualified for whatever you're applying to. In order of decreasing impact:
then all my letter can say is your grade in the class, and that's something your transcript already does. I also only want to recommend students who are friendly, polite, and a pleasure to be around.
- if we never did a project or reading group together,
- if you never came to office hours and had an academic conversation with me,
- if you did not actively participate regularly (i.e., almost every day) in class,
- (For certain applications:) Be a mathematics major. Especially if your application is math or STEM focused, letters are better received when they come from professors in your major/field. At the very least, even if you're not a mathematician, it'd be good if you've taken more than one class with me.
- Make sure I'm an appropriate person to write your letter. Note that I'm currently still a grad student, so my letter carries very little weight in academic applications, like to grad school, to med school, for scholarships, etc. One of the purposes of a letter is to attest to your ability to succeed in the program you're applying to, so in reviewers' eyes, if I haven't yet succeeded in my own graduate studies, then my opinion on your ability to do so is worth a lot less than, say, a professor's.
- Be in good academic standing. On principle I will not recommend a student who has been found in violation of academic integrity. If you have ever "grade-grubbed" or asked for class policies not to apply to you, I refuse to recommend you. You need to have passed all your classes.
- Not apply to the following kinds of organizations.
I am happy to have an in-person discussion about my reasons for this stance. I've adopted this list nearly verbatim from Matthew Gentry Durham's, itself adapted from the Just Mathematics Collective's.
- The military industrial complex.
- Intelligence agencies.
- Policing and the prison industrial complex.
- Any organization that relies heavily on insecure employment, poor labor conditions, and/or union-busting.
- Any organization whose existence requires destructive environmental practices.
If you don't meet all of the above conditions but believe you have a compelling reason why I should write you a letter anyways, let me know
! I'm happy to hear you out and I've written letters before for strong students who may not meet every condition above. In your email, please indicate that you've read everything here and explain why you deserve an exception. I am uncompromising on my ethical concerns but typically reasonably flexible otherwise. Knowing you well trumps almost everything else.
If you do meet every condition listed above, please email me
to ask for a letter. Here's what I'll need from you:
- Your CV. CV stands for curriculum vitae (literally, the course of your life). In academia, unlike most other disciplines, we use CV to mean a document that displays your complete academic and professional history, as opposed to a résumé, which is a summarized version, focused more on your professional history than your academic, that you might use to apply to a (non-academic) job.
- Your transcript. Unofficial version is fine. Please don't spend money on my behalf.
- Your performance in my classes. Remind me which classes they were, what semesters they were held, and what grades you earned (or just point them out when you send your transcript). If there were any projects or major assignments in our courses, resend them to me.
- Information about the application. What are you applying to? Do they have a program announcement? How do I submit my letter? Are there multiple different programs (e.g., applying to several grad schools)?
- Why you're applying. What are your career and life goals? How does this application further those?
- Your parts of the application. This may include things like your statement of purpose, your research summary, etc; I want everything. I need these to be PDF. (I don't know what the application requires, but they probably want PDF too.)
- The names and details of your other letter writers. Why did you pick them to write letters? How can I complement their perspectives?
- Information that you suggest I put in my letter. I'll ultimately decide what I write, but are there particular accomplishments or qualities about you that you'd like me to discuss? In particular, please share with me at least two or three personal anecdotes about our relationship (even if you don't specifically want them in the letter); this helps jog my memory and highlights to me the things you've valued about our relationship and why you'd like a letter from me in particular. (Note that to do this well, you need to have made an impression and sought a connection with me, like I mentioned above!)
- My due date. It needs to be a long way away, in order to give me ample time to write your letter well. I'm not going to set a time frame in stone, but at least over a month is really preferred. If it's too soon I'll say no. It's also YOUR responsibility to keep reminding me about the deadline. You have my permission to pester me incessantly when it gets close.
Don't worry; most of this stuff you'll likely need for the application itself anyways. Go ahead and send all this info at the same time as when you ask for my letter. I'll likely need to see it before I can commit to "yes" or "no," this avoids too much back-and-forth, and even if I end up unable to write your letter, it benefits you to have all this information together in one place. Your other letter writers will probably appreciate that you've gathered it too.
You will also need to waive your right to view my letter. This is a cultural norm that may seem odd at first, so let me explain. It is standard, especially in academia, for applicants to waive the right to view their letters. You are not the target audience for your letter, and often a candid letter assesses you honestly by making direct comparisons to your peers; for example, see this sample letter
by Evan Chen. This is an unequivocally positive endorsement, but it would not be appropriate for you to read such a letter about yourself. Another significant reason is that when you don't waive your right, the people who review your application will sometimes (fairly or not) view your letters with skepticism. I do not want to risk your application being viewed less favorably, the reviewers finding your actions gauche, or my letter coming across as insincere or not candid. Thus I will not write a letter if you can view it. Please also note that waiving your right does not mean your letter writers will now talk poorly about you or sabotage your application! I only agree to write letters at all if I can write a strong one. If you don't trust a professor to have your best interests at heart when you waive your right to view, you probably don't have a relationship that warrants a letter anyways.
After you hear back about your application, please let me know how it went! I love celebrating my students' successes with them. 🙂
This might all seem like a lot, but it's pretty par for the course. You can see similar advice from Tarik Aougab
, Adam Boocher
, Evan Chen
, Keith Conrad
, Taylor Dupuy
, Matthew Gentry Durham
, Megumi Harada
, Leo Herr
, Chris Kapulkin
, Gizem Karaali
, Kiko Kawamura
, Kiran Kedlaya
, Michael Orrison
, Rob Pollack
, Bjorn Poonen
, Dhruv Ranganathan
, Joe Silverman
, Mark Tomforde
, Ravi Vakil
, David Zureick-Brown
, and I'm sure 1000 other people whose resources guided me through the norms of academia.