LOR From a Research Internship Supervisor

LOR From a Research Internship Supervisor

Yes, a research internship supervisor can write an MS or PhD letter, and it counts as academic. Cambridge lists "Your tutor from an academic internship at your current or another institution" as an academic referee. Toronto CS wants at least two letters that "speak to your ability to conduct independent research". The letter works only if it stays inside the internship: the dates, one project, what the student did alone, and a comparison with the other interns.

This page is for the host supervisor writing that letter, whether the summer was in an IIT lab, a Mitacs Globalink placement in Canada, or a European research group. The student's side of the same choice, which letter to ask for and how to write the summer into the SOP, is in the guide to a summer research internship in your SOP. For a free Quick Review of your draft, choose the graduate recommendation letter rubric.

Every university and Mitacs page quoted below was checked on October 31, 2026. Reference rules are set per course, so the student should send you the exact programs they are applying to.

Does an internship supervisor count as academic?

Our view: for a research MS or a PhD, a host supervisor who watched the student do research for eight weeks can carry more weight than a home professor who taught one large course. The basis is the table: Toronto and UIUC ask for research ability, and a course grade does not show it. The countercase: for a taught master's that asks mainly about academic performance, the home professor who knows the whole transcript may be the better writer, and Cambridge requires one letter from the current or most recent degree anyway.

What a short-contact letter can credibly say

The forms tell you what you will be asked. Sydney's referee form opens with "In what capacity, how well, and for how long, have you known the applicant?" and asks you to "compare this student with others known to you who have been accepted in previous years or are now applying for admission." Cambridge's guidance for academic referees asks for your role "and how long that was for", and for "the size of the group against which you are comparing the student."

If eight weeks gave you nothing you could name, decline writing the letter early so the student can ask someone else.

Eight weeks is a short window. Stay inside it.

If a form asks about something you never saw, such as teaching, use the form's "Unable to judge" or no-basis option rather than a guess. Our guide to no basis for judgment ratings covers that choice.

Our recommendation: put the dates in the first sentence. A reader who sees "June 3 to July 29, 2026" adjusts every later claim to that window, and the letter is more believable for it.

Constructed sample letter from a summer host supervisor

Constructed example, written by us, not a real recommendation letter. Context (invented): Dr. S. Menon hosts eight summer interns a year at an invented institute; the student, Ananya Rao, is applying to US PhD programs in computational biology.

Dear Admissions Committee,

I supervised Ananya Rao as a summer research intern in my group at the Coastal Genomics Institute from June 3 to July 29, 2026. I met her twice a week and read every analysis she produced. My view of her is limited to those eight weeks and to one project, and everything below stays inside that window.

Ananya's project asked whether a published marker set could identify two closely related algae species in our field samples. I gave her the pipeline and the data. In week three the pipeline assigned four of our field samples to a species never recorded on this coast. Without being asked, she pulled the reference sequences behind those calls, found that one entry in the public reference database carried the wrong species name, and confirmed it by running the samples against a second, independently curated reference. She showed me the check, the evidence and the corrected result, and she drafted a correction request to the database maintainers. The group now checks every new marker set against two references.

In the last three weeks she rewrote the analysis so another student could run it from one command, and she presented the corrected result at our group meeting. Her explanation of why the original call was wrong was clear enough that a visiting colleague asked for her notes.

I have hosted 46 summer interns over the past six years, and that is the group I am comparing her with. She is one of the two strongest of the 46. What sets her apart is not speed. It is that she distrusted a surprising result and tested it before anyone asked.

I did not see her coursework, and I cannot comment on her teaching. Her written English in our reports was clear and precise.

I would accept Ananya into my group as a PhD student without hesitation, and I recommend her strongly for doctoral study in computational biology.

Dr. S. Menon Senior Scientist, Coastal Genomics Institute

That letter is 331 words. Its people, institute and numbers are invented for the example, not facts about any real lab.

What it does, in rubric terms:

• Basis first. Dates, meeting frequency and the limit ("stays inside that window") come before any praise.
• One incident carries the letter. The second-reference check shows independence, method and care with sources without the words "independent" or "careful".
• A comparison with its group. "One of the two strongest of the 46" names the group and the window, the way Cambridge's question asks.
• Limits stated plainly. No coursework, no teaching. That makes the rest more believable.
• An endorsement a committee can quote. "I would accept Ananya into my group as a PhD student" is the strongest sentence a host supervisor can write, and Cambridge asks for it directly: "whether you would want to supervise the student yourself".

Home advisor vs host supervisor

Where a program takes three letters, the student can send both. The home advisor covers the degree, the transcript and the thesis; you cover the summer. Do not repeat each other. The student's side of this choice is in the summer research SOP guide, and the case where the home thesis adviser cannot write is in PhD references without your thesis adviser.

If a postdoc or PhD student supervised the intern day to day, Cambridge's rule matters: "Do not nominate PhD students." Its referee guidance suggests the faculty referee talk to them instead: "if the student was taught by a PhD candidate, they may be a useful source of information." For a postdoc, the options are covered in joint PI and postdoc letters.

Mitacs and exchange-semester supervisors

Mitacs Globalink. A Mitacs host professor in Canada can write an ordinary admissions reference. That is a different document from the Globalink Graduate Fellowship letter, which only confirms supervision. Mitacs says the fellowship "will officially cease to be offered by 2025 and will, therefore, not be available to GRI 2025 interns" (Mitacs Globalink Graduate Fellowship). The Canadian reference letters guide covers the other Mitacs rules.

If you only taught the intern in a course, the guide to a letter for a student you barely know fits better.

Exchange semester. Cambridge defines an academic referee as "someone who has guided and assessed your academic work at a university or other higher education institution." A professor who supervised a project during an exchange semester meets that. A professor who only gave a grade in a large lecture course has little to add to the transcript.

Before you submit the letter

1. The dates of the internship are in the first two sentences.
2. The project is named in plain words, with the student's part separated from the group's.
3. At least one incident shows something the student did without being told.
4. The comparison names the group (interns, how many, how many years).
5. Anything you did not see is stated as outside your view, not guessed.
6. The endorsement says what you would do: take them back, supervise them, recommend them "strongly".
7. You used your institutional email. Toronto CS says "Gmail addresses are not acceptable."
8. You wrote the letter yourself. Stanford: "Drafting, writing, translating, or submitting your own reference, even if asked to do so by a recommender, is a violation of our application terms."

Questions about letters from internship supervisors

Is an eight-week internship long enough for a recommendation letter?

Yes, if the letter claims only what eight weeks showed. Cambridge names the internship tutor as an academic referee, and the forms ask how long you knew the student, so a short window stated honestly is expected, not penalized by any rule we found.

Should the letter mention the internship program by name?

Name it if it helps the reader place the work, such as a competitive summer program. The project and the student's actions matter more than the program's name.

The student wants to send me a draft. Should I accept it?

No. Ask for a short list of facts instead: dates, project title, the student's outputs and the programs. Stanford calls an applicant drafting their own reference "a violation of our application terms". The student's side of that request is covered in how to ask for a grad school letter of recommendation.

Can I reuse one letter for several programs?

You can reuse the core, but tailor the closing to each program. Cambridge requires "a separate reference for each course", and says "We will not transfer references from any previous application." If you supervised the student's thesis rather than an internship, see the PhD recommendation letter sample from a thesis supervisor.

If you are still choosing between this supervisor, a professor and your HoD, see who should write your MS LOR.

Get a free review of the letter

Get a free Quick Review of your recommendation letter against GradPilot's graduate school recommendation letter rubric. It reads how you know the candidate, incidents instead of adjectives, comparisons that name their group, and the strength of your endorsement. Writing for a UK, European or Australian program? The international academic recommendation letter rubric also checks that your comparison makes sense to a reader outside your system.

More guides for recommenders and international applicants are in the graduate school essays hub. For what PhD committees read for across all three letters, see PhD letters of recommendation.

Sources

Retrieved October 31, 2026.

• University of Cambridge, References to support your Cambridge application (no cycle stated): academic referee examples including the internship tutor; one reference from the most recent degree; no PhD students; separate reference per course; no transfer of references.
• University of Cambridge, Guidance for academic referees (no cycle stated): role and duration, comparison-group size, "whether you would want to supervise the student yourself", consulting PhD candidates.
• University of Toronto, Department of Computer Science, How to apply (Fall 2027 applications): two research letters, preference for research referees, no Gmail.
• University of Illinois Siebel School, Application process and requirements (no cycle stated): letters to address research abilities for MS and PhD.
• University of Sydney, Postgraduate Research Referee Form (form version vMay16): capacity, how well and how long; comparison with others; "Unable to judge".
• Stanford Graduate Admissions, Recommendations (no cycle stated): "know you well"; sole-author rule.
• Mitacs, Globalink Graduate Fellowship (no cycle stated): the fellowship ends and is not available to GRI 2025 interns.