I am pursuing a Masters degree in pure math, and will be applying to grad schools in the USA and Europe (and some other places) for a PhD. I have one letter confirmed from my thesis work, which I believe is my strongest and will be my primary letter. I hope to have another from a course that I'm taking right now, albeit not a very strong one in my opinion because we have started communication just a few months back and the letter needs to be submitted before the evaluation of his course.

More concerningly, a third letter is up in the air. I could approach a professor whose course I have done decent in (but not great), or a professor at my earlier college who I have done great under, but whose collaborations/connections are not very notable/have not spread far.

Another option that came to mind is this person who has been involved in my math journey for quite some time, but is now in an entrepreneurial position in a moderately-established startup organisation to train people in math (olympiad/advanced), not at any college/university. He himself completed his PhD (however in a pure math field different from mine) a few years back from a reputed university from the USA, is actively researching and has had his work featured in and himself invited as a speaker to quite a few conferences. He is someone I have discussed a lot of math with (including a bit of budding research in his area) and knows me well personally, hence could be able to write a much less generic and more comprehensively personal recommendation.

Given the choice, which ones should optimally be my submitted letters and in what order?

2 Answers 2


This applies to the US and possibly elsewhere. No, the letters don't need to be from professors at universities. However, the best letters, the ones that will matter, are the ones that both confidently predict your success in graduate study and can be relied on for the knowledge of the writer. That means someone who both knows you well and also knows about graduate study. This is more critical for doctoral study than at the masters level, but still applies.

I'd guess that your "optional" writer is an appropriate candidate for the above, and would be even for doctoral study. It isn't about them, but about what they can say about you and be trusted in its accuracy. I don't see any problems here.

  • 8
    Can anecdotally confirm from the student side: By the time I applied for my PhD, I was 15 years into US industry and largely out of touch with my previous academic mentors-- enough so that it felt weird hitting them up for recommendations. My letters were all from industry, one with a PhD, two without (but one responsible for managing PhD-level people.) I got in.
    – Anonymous
    Oct 15, 2023 at 21:29
  • Thank you very much.
    – Anh Nguyen
    Oct 18, 2023 at 13:33

As a general rule, it is perfectly acceptable to use letters of reference from successful people in research and industry positions outside academia. Some of these people have equivalent academic stature to professors in a university and have similar backgrounds in their education and research records. You should seek letters from people who can comment positively and credibly on your research work/potential. This need not be restricted to university professors.

  • Thank you, I'll keep that in mind.
    – Anh Nguyen
    Oct 18, 2023 at 13:33

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .