Some career advice for my juniors

This blog post is a consolidation of the recurring advice I have given to many juniors who have asked me about finding internships and research opportunities. I don’t know if my advice helped them, but I do hope it will be of use to others¹. This is highly opinionated, and I am sure there are other perspectives that one should consider based on their specific circumstances and goals. While I have a narrow range of personal experience to draw on, I often rely on the advice that I have received from my mentors and also the anecdotal experiences of my friends and colleagues. As you always should, take this with a grain of salt.

This blog follows the ChatGPT-interview format that I first used last year². Like I did then, I have only lightly edited ChatGPT’s responses and have left my own nearly verbatim.

Host (ChatGPT): Every year, especially before summer, many undergraduate students find themselves asking some version of the same question, “What should I do this summer?” The question may be about internships, professors to work with, research topics to explore, whether to apply to programs abroad, or even whether one should be thinking about academia or industry at all. But beneath these concrete decisions often lies a deeper uncertainty about how one ought to think about opportunities in the first place. In this conversation, Mrigank argues for thinking in terms of experiences, mentorship, transferable skills, and alignment with longer-term goals—and for being deliberate, even a little selfish, in choosing where to spend one’s limited time and energy.

---

Host: Mrigank, when juniors come to you unsure about which internships, research opportunities, or topics they should focus on, how do you help them start figuring out what direction to take?

Mrigank: I always begin by asking what they want from their bachelor’s degree, that is, their final goal. Broadly, most people fall into two categories: those who want to do research, which often means pursuing a PhD, and those who want to leave academia and join the industry. Of course, this is only a broad split, and most students are either torn between these two paths or have already chosen one over the other.

For students who want to end up in industry after their undergrad, I admit that I don’t really have any expertise to offer, because I did not participate in the campus recruitment process at my university. What I know from friends is that you need to practice solving LeetCode problems, prepare for interview questions, and maybe build some small projects to put on your CV. If you worked on some research project, it might be a bonus. Some of the things I say about opportunities and experiences might be relevant to such students, but I am not sure how much.

For the students leaning towards research, I have a little more to say.

---

Host: Since the path one ultimately wants to take can shape what kinds of opportunities one should look for, what would you say to someone who does not yet know whether they want to pursue research or jobs?

Mrigank: This is a common situation, particularly in universities like mine, where both paths have been well-trodden. I was almost certain that I wanted to do research, so this was not a dilemma for me. But I think sampling both directions is a good idea, especially in the freshman and sophomore years. One can try out a research project with a professor, even if it’s not in an area they are passionate about, in order to get a sense of what research is like. As I mentioned before, it will be a worthwhile experience to have on your CV even if you later decide to go into industry. If you don’t like it a lot, you can then try an industry internship the following summer.

Alternatively, if you have a long-term vision for yourself, you can think which path is more likely to lead you there. For example, some of my juniors have told me that they want to work at DeepMind³. For these students, I pointed out that most of the people doing most of the cool stuff over there have PhDs, so pursuing a PhD would be a statistically better choice. Actually let me point out a common misconception here. Many undergrads think that a PhD is for people who want to be professors, but this is simply impossible⁴ because manyfold more people graduate with PhDs every year than the number of faculty positions available!

At some time during one’s undergrad, it is important to get clarity on which path one wants to pursue. While one should indeed be open to exploring, the reality is that undergrads have limited time and resources, which have to be utilized with some planning before stepping into the next phase of their careers. Since applications for both graduate school and jobs start roughly a year before their start date, it helps to have a clear direction by the middle of the junior year⁵ as the two tracks demand different kinds of preparation.

---

Host: Students leaning toward research often worry about finding the right topic to work on. Do you think that is the right thing to be worrying about?

Mrigank: The right topic is the one you like! I know that insight does not help much because there are so many topics out there, and it is hard to know which one you will like until you try it. Sometimes you may end up liking multiple topics too! But ultimately you should hope to narrow down to one so that you can build depth and contribute something meaningful to it. I think with the limited time that undergrads have, sticking to one topic is a good idea.

But what topic? If you don’t know, I will suggest that you should not worry about it too much. My advice is to think in terms of experiences rather than topics when it comes to finding or choosing opportunities. The key here is to gain the transferable skills that will be useful in a future research career. These include studying a problem by reading papers, thinking critically about existing work, formulating your own questions, designing and running experiments, building software systems, writing papers, and so on. Undergrads are a blank slate and can enjoy anything they put their mind and heart into, so the exact topic is not as important as the kind of experience you get from working on it. Surely, it is helpful to have preferences which you might build from taking classes or working on projects. For example, I ruled out transit network design after a short internship I did in my first summer. Similarly, I had a liking for applied areas and did not want to work on theory. If you have some preferences, you can restrict your search to those areas.

It is a little unsettling for many students, and understandably so, to have to decide on a topic so early in their careers and also so quickly. But the reality is that you are not really locking in! For example, as far as I have seen, it is typical for people to begin their PhD with a topic different from what they worked on in their undergrad. In fact, it is not even uncommon for PhD students to switch topics during their PhD. Professors pick up new research areas all the time as well. So there is plenty of time to explore in the future.

The bottom line is that you should pick up opportunities based on the experiences you will obtain from them, not just the specific topic you will work on. If a good opportunity comes along, go for it and stick to it for a while!

---

Host: If transferable skills and experiences are the main thing to optimize for, how should one decide which research opportunities are worth pursuing?

Mrigank: As I said, any opportunity within the space of topics acceptable to you and that provides you with the experiences I mentioned earlier is worth pursuing. I understand that there is some disparity between different universities when it comes to how accessible such opportunities are, but one should at least be mindful of what they are looking for. And these opportunities can come from anywhere and from anybody—from a professor in your own university or from a professor abroad, from a senior professor or from a junior professor, or from an industrial research lab, and so on. While it is true that there is some correlation between the reputation of the professor or the organization and the quality of the experience you will get, some students tend to sweat over it more than necessary. I think many of these notions are not generalizable and very specific to the mentor and the project you will work on. They are also somewhat secondary to the other factors I mentioned. Not only should you know what you want from an opportunity, I believe you should also be upfront about it with your potential mentor. Discussing your goals and expectations can help them help you better, and it also conveys that you are serious about your work because you have a vision for yourself. For example, your goal may be to publish a paper or to get a strong letter of recommendation, and your expectations may be to have regular meetings. On the flip side, if you are unsure your expectations will be met, you can toss out the opportunity and look for another one. You can also speak to other people who have worked with your potential mentor before to get a sense of what working with them will be like and whether their mentoring style is a good fit for you.

My mentors have played a huge role in my journey and I cannot stress enough how important it is to find good mentors. The experiences I mentioned earlier, related to the research process, will probably be handed down to you by a good mentor. But you will have to be a good mentee to extract the meta-level knowledge that an experienced researcher has to offer. By being proactive and making the most of your interactions with your mentor, by asking questions and carefully listening to them, you can learn how they think about research, how they break down complicated problems into actionable steps, how they present their ideas, or how they critique other researchers’ work.

---

Host: We’ve talked about what to look for in an opportunity. Are there common pitfalls or red flags students should be mindful of, even when something seems appealing at first?

Mrigank: I do have a few things to say about this. I can think of three axes that I would like to touch upon. Please keep in mind that I am not making general statements here, and these are specific cases that students should look out for.

The first is alignment. If an opportunity does not align with your goals, it is probably not worth pursuing. For example, if you want to apply for PhD programs, you should probably not spend your summer doing a product- or service-oriented internship in a company. There might be peer pressure, or you might get a lucrative offer from a popular company, but you should keep in mind that you are spending time away from potential research experiences that could have helped you build a stronger profile for graduate school applications. There is no correct answer here, and I am simply asking you to weigh your options carefully.

The second is exploitation. You should never let yourself be exploited under the guise of an opportunity. I can tell you that there are many startups that will tell you, particularly if you’re a student in a well-known university, that they are building something that will change the world and you will be working on the bleeding edge of research if you join them, but will make you perform grunt work for 12 hours a day without proper compensation, mentorship, or learning. Unfortunately, this is far more common than you might think, and while these startups themselves might be legit, they know that undergrads are energetic and hardworking, and take advantage of their eagerness to prove themselves. Not all startups are like this, but many are, and if you find yourself in such a situation, just run. It is simply not worth your time and talent.

The third is prestige. I will give a very specific example for this: foreign research internships. This is a big thing in my university, perhaps because historically many students from my university have visited institutions abroad for research internships. Lately it has become somewhat harder to get into such internships because of the suspension of several programs, but the buzz around them carries on and many students stretch themselves thin trying to find such opportunities. Now I do agree that the majority of students who participate in such internships benefit from them in terms of their immediate career goals, for example, by working on a meaningful project, publishing a paper, or getting strong letters of recommendation. But a sizeable number of students end up with an experience that is not really worth the hassle of visa applications, travel logistics, and possibly even funding their own visit. For these students, the internship is too short to get any tangible work done and there is no plan to take the project forward after the internship. This is particularly the case when you cannot work remotely, like in many non-CS areas. Essentially, foreign internships are not a golden ticket, and it is likely that one can benefit just as much or even more from working with a professor in their own campus. At the end of the day, it is the quality of the experience that matters more than the location where it takes place. So while I will not discourage students from looking into such internships, I will suggest, as always, to be mindful and deliberate in what they are looking for from them.

---

Host: If you had to distill all of this into one guiding principle for choosing opportunities and making these decisions, what would it be?

I think the underlying principle is to be deliberate, might I even say selfish, in deciding where to spend your time and mental cycles. You should know what you want from what you do, and how you could get it. You should be fearless in pursuing your wants, in asking for help if things are not working out, and walking out if there is no reconciliation with your goals.

---

Host (ChatGPT): Perhaps the central idea in this conversation is that undergraduates should think less in terms of chasing the right opportunities and more in terms of seeking the right experiences. In a stage of one’s career where time is limited and exploration matters, that is an important distinction to keep in mind.