How to get an internship at Harvard/Stanford/MIT or similar university

If you are in college, doing an internship is the best way to spend the vacation and add a major point in your resume. Most students do internships close to their college or home, some even do in far off cities. But very few do a university internship abroad. For a university, they are always looking for skilled people with an aptitude for a very specific niche.

If you want to be one of those who get the chance to go abroad and pursue an internship, read on.

In short the process is: you identify the field you want to work in, find out the professors working in the area (s). Find out their email id, and send each of them a customised email. You will have to contact each professor individually, and request him to hire you.

You need to be very careful in your approach, and please avoid the most common mistakes that people make in this process, which I’ve detailed later in this article.

Note that if you are going to be working in a university, you will be pursuing a research internship, and in most cases it will be on a project the professor is already working on. Very rarely any Professor would start a new project for an intern.

The wrong approach:
1. Go to a university’s website
2. Make a list of Professors and email ids
3. Send same email to all of them asking for internship

The correct approach:
1. Go to
2. Search for your area of interest
3. Read the summary of papers and make a list of the authors whose papers you liked. If email id is not mentioned in the paper itself (or if you are unable to see the paper then search the authors’ name and affiliated institution on Google and get their email id.
4. Write a customised email for that professor. (When you write a mail to someone who you don’t know, it’s called a cold email). Make sure it has the major elements of a cold email.

Sample email

(Andrew Ng is a very famous Machine Learning researcher from Stanford University, who also founded Coursera)

Subject: Regarding possible internship opportunities under you

Respected Prof. Andrew,
Greetings! I am Bhavesh, a fourth year undergraduate student pursuing from XYZ University. I am writing this mail to enquire about possible internship opportunities under you. [1. This is where you give simple introduction of yourself and in a straightforward manner, ask what you want from the person]

While researching, I stumbled upon and went through your paper “Rectifier Nonlinearities Improve Neural Network Acoustic Models” and really liked the work you and your team did. I am also very interested in this topic and have been working on it for quite some time. [2. This is very important, it shows that you are writing a personal mail and not one which has been sent to multiple people. It also shows that you are actually interested in the professor’s research areas.]

In the past I have worked on multiple topics in applied Machine Learning, most recently being under Prof. K. Venugopal, XYZ University, on the topic “Using Neural Networks for Stock Market Index Prediction”. Our work was well appreciated and also presented at the ABC Conference. [3. Here you are trying to tell that you have some experience in the field where you want to work.]

I went through several other research papers of yours and really liked your work. It would be an honour for me to work under you.

Please find my resume attached. [4. Your resume is required, Prof. would want to see more details about your education and experience before he takes the discussion forward.]

Waiting for a positive reply.


XYZ University

Common mistakes:
Sending the same mail to multiple professors
Not talking about your research experiences and why you are interested in his
Starting the approach very late. Ideally you should be done at least 2-3 months before the expected start date of your internship.

Hope the above article helped. Reply to this, if you want any questions.


No comments yet. Why don’t you start the discussion?

Leave a Reply

Your email address will not be published. Required fields are marked *