Grad School Application Tips from a Swat CS alum

This is a ~little~ rather long guide containing some advice on applying to graduate schools, based on the experiences I had in Fall 2019 - Spring 2020. I hope that this will serve as some words of wisdom, and also as some encouragement. While this was originally intended for Swarthmore CS students, I thought it might be nice if it was publicly available in the case that someone else would also benefit from it. Please read the disclaimers first!! Also, please feel free to email me ayaka@cs.cornell.edu if you have any questions or concerns, I'd be happy to help!

Disclaimers and About Me

Disclaimers

EVERYONE’s grad school application experiences are different. Please take most things in this document with a grain of salt, because this doc is based on my own experience. That being said, I will try my best to keep this doc as general as possible.
Adding onto #1, the expectations for grad school applications vary based on field, even sub-fields of CS. Also for MS vs PhD.
I only applied for CS PhD programs, so that’s what I will be talking about. Hopefully some of the things in this doc (ex. GRE stuff) will be helpful for people in other situations though!
I do not know what got me into grad school, so please don’t expect any info of the sort in this doc. As everyone says, grad school applications are a crapshoot. Sometimes a good one, sometimes a bad one. The goal of this doc is to feature things that I wish I knew when I was applying. (I have to admit that most of this is advice you can hear from your profs)

Given the Disclaimers, I should probably explain a bit of my background...

About Me

I am currently a first-year PhD student at Cornell University(as of fall 2020). At Swarthmore, I was a double major in Computer Science and Music.
I applied in my senior fall, so the ideal target audience of this doc might be people who are thinking of doing the same thing.
My main area of interest when I applied was Programming Language theory and applications. While I am still interested in PL theory and applications, I've pivoted ever so slightly since I started grad school; now I mainly work in Software Engineering, where I get to apply concepts from PL theory.
I did PL research during my Junior spring, the summer between my Junior and Senior year, and all throughout my senior year.

General Advice

The big goal of grad school applications is to get into at least one (1) school, and at most enough numbers of schools so that you can find one (1) school that works best for you.
It’s totally OK not to know what you want to specifically work on by the time you apply, or even by the time you get into grad school!! I’m still not quite sure what I want to do specifically.
- That being said, it’s always good to have a general idea (even though that might change!) and have some buzz words by your side when someone asks you what you are interested in.
According to the wonderful Swat CS profs, you should apply to 9~12 schools. I full heartedly agree, and here are a couple of notes on this…
- I applied to 15 schools and it was definitely way too much.
- Yes, increasing the number of schools would increase your chances of achieving the big goal (refer to top of this section).
- BUT ask yourself, “is this extra school somewhere I’d definitely want to go if I was accepted?”
- Chances are that you’ll get accepted to one of the 9~12 schools you apply to :) If you want to apply for more schools, just make sure to acknowledge that adding another school means adding on the number of hours you have to work on apps!!
Your research/academic advisor is an amazing and wonderful resource! Check in with them about the little decisions that you’re making (like, what schools you are applying to), and ask them for advice!! Start talking to them early!
Make sure to keep a personal agenda for how you want application season to play out, and pace yourself as you go!! As hard as it is to beat the procrastinator spirit, it's really helpful if you push through all of this gradually.
Ultimately, it probably doesn’t (and shouldn’t) matter when you submit your application, as long as you have it within the application submission deadline.
- That being said, something that was helpful for me was to set an early personal deadline for submitting my first application. (Around the beginning of November... but in reality it didn't even need to be that early.)
- This was especially helpful for me because it meant that I needed to work on the core of my application (Personal Statement, CV, etc) before that date, and all I had left was filling in more school-specific things.
Most CS PhD programs should provide full funding for the whole time you are there. DON’T APPLY TO A SCHOOL THAT DOESN’T PROVIDE FUNDING!! You need to live!!

Choosing Schools to Apply to

If you are applying CS and haven’t already checked out csrankings.org, you should!! The rankings are based on papers (I think?) and some info is a bit inaccurate but it’s a great way to figure out what schools you might want to apply to.
Ask your research advisor/prof who works in the research area what schools are good!! They’ll be super helpful.
Another nice venue to check out are the various academic conferences that relate to your area of interest! (Again, ask your research advisor/prof who works in the research area for what conferences you should check out)
It seems like CS PhD programs in Canada/Europe recommend you get a masters first? But I might be wrong on that.
Make sure to apply to schools that you genuinely think you'd like to go if you were accepted!!
- One thing that could help is looking at the various research groups, and the webpages of the students in the school!! If you want to be like them, that's pretty awesome.
- Don't forget to think about geographical location too! Are you willing to move to the specific location that the school is at?
Make sure to apply to a wide array of schools (“reach” school, “this school seems reasonable" school, “safety” school) placed in various places on the rankings.
- You never know what would happen!! The reason I have the categories in quotes is because what you perceive to be the probability of you getting into the school is NOT the actual probability of you getting into the school.
- Don’t give up on a school just because it seems like you won’t get in. You don’t know, you might just get in!
- That being said, aiming far and wide will increase the probability that you will be accepted somewhere :)

GRE

If you’re looking for a target score, UPenn CIS has some GRE score statistics. Some other schools post their target scores too, so it’s worth looking them up.
All of the writing prompts are featured on the ETS website!! The issue topics are here, and the argument topics are here.
ETS "sells" you two practice tests for free when you sign up for the GRE. You should definitely “buy” and take them!! They help for 1) going through the GRE experience and 2) practicing using their actual interface!
Do a lot of practice problems, and go through them to see what you got wrong and why! The 5lb book of practice problems by Manhattan Review was helpful for me, but any book that has a lot of practice problems would be good.
If you’re struggling on vocab (and motivating yourself to learn vocab), setting a low number of words to remember every day, and reinforcing them every weekend, is pretty helpful.
If you are doing summer research during the summer after junior year and are intending to apply in the fall, taking the ~2 weeks between the end of research period and the start of the semester is a perfect time to hunker down on the GRE, take it, and get it out of the way!!
Unless you really think you can significantly improve, don’t take the GRE more than twice… it really isn’t worth the money and effort after a while. When you take the GRE, they allow you to send your scores free of charge to up to four institutions. Take advantage of this!! Sending GRE scores is pretty expensive :(
Your goal while taking the GRE should be to get a good enough score that you won’t be weeded out just because of your GRE score. A good enough GRE score is perfect!!
MAKE SURE THAT YOU ARE SENDING YOUR GRE SCORES TO THE RIGHT GRE SCORE SUBMISSION CODE FOR YOUR PROGRAM!!! I can’t stress this enough

Planning Fall Semester

This section is admittedly pretty Swarthmore specific, but I hope the sentiment that YOU SHOULD MAKE SURE NOT TO OVERLOAD YOURSELF would be conveyed to everyone :)
Applying to grad schools is basically like taking another Swat course. (maybe even a 1.5 credit course) It will take a lot of mental energy and time away from you, so please bear that in mind.
- Even after you put together your statement of purpose, just filling out the random information on the application takes ~2hrs on avg for each school.
- This is NOT the time to be taking 5 credits. Please.
Pass-fail is your friend!! If you still have them.
Many of my Swat professors were very generous and supportive, so if you ever feel overwhelmed about balancing coursework and applications you should try telling them that you are applying to grad school. That being said, you don’t want to be taking advantage of this too much.
Be sure to still make time to hang out with your friends and have fun once in a while!! This process is draining, but you need to not let it take over your entire life :) Supportive friends are amazing!!
Overall, this is gonna be a rough time and one of the best things you can do is to take care of yourself and be generous to yourself! There will be days when you can’t achieve as much as you wanted to and that’s ok :)

Statement of Purpose

A lot of statements of purposes look very different, but one thing to keep in mind is that you want to subtly introduce a narrative but without getting too anecdotal. Demonstrate what you’ve done, what your motivations for going to grad school are, your goals after grad school, and your contributions to academia (ideally you would have some research experience by this point) in a concise way.
- What in your background motivates you to work in the field/area that you are interested in going into? Why did you choose this field?
- What in your background demonstrates your potential to work on hard problems in your field?
- What kinds of experience do you have in research/tackling hard problems in your field? What did you find fun or exciting while working on these problems?
- What kinds of other experience do you have that might help you as an academic? Have you done any TA-ing or tutoring?
- As everyone says... show, don't tell!!
BE. BOLD. BE. ASSERTIVE. This statement is your chance to shine!! You are most likely a wonderful computer scientist and even if you feel insecure about your level of expertise in the subject or in CS, you can't let that show. If you let it show, it just gives your audience a chance to underestimate your potential.
- Believe in yourself!!
- If you are having difficulties in believing in yourself (because I also often have a hard time with confidence issues), at least pretend to believe in yourself.
- If that's also kinda hard, "Believe in the me that believes in you!!" (imaginary bonus points if you get the reference, please let me know through email for the sake of laughs and giggles.)
Make sure to be as direct as possible!! You want to get straight to the point.
Having a little madlibs section to talk about specific schools is very helpful!
- The part where you talk about the specific school doesn’t need to be super blown up (at least for CS). It’s best if you can explain why you’re applying and why you’d go there if you were accepted shortly and concisely!!
If you have published a paper or submitted a paper to a conference, or have any other kind of publications, you should include a short sentence about that (after giving a high-level explanation about the research you did that lead up to that paper, of course).
The WAs are absolutely wonderful and you should ask them to take a look at your Statement of Purpose draft for general grammatical and flow purposes. There’s a big chance they’re not a CS person or they don’t know what CS Statement of Purposes are supposed to be like, but they are a great resource either way :)
Lots of people have different opinions on what a Statement of Purpose should look and sound like. Sometimes you’ll hear contradicting opinions, but that’s just how it is. Don’t ask too many people, because it could muk the statement up a bit.
As you make edits, you’ll get to the point where you can continue making edits forever and it really won’t change the overall effectiveness of the statement. Definitely put in a lot of thought into the Statement, but don’t overpush it!!

Letters of Recommendation

Letters of Recommendation are one of the most important parts of your application!!
Make sure to ask your professors pretty far in advance (during September/early October) - they have a lot of things on their hands too!!
Definitely definitely ask the professor you’ve been doing research with to write you a letter of recommendation. The more that the prof can talk about your research efforts the better!
Be sure to ask professors who are in a similar area of your intended research field/area.
- If you’re going to apply to CS programs, professors in the areas CS, Math/Stat, Engineering, Econ would be good too!
If you are intending on applying after you graduate from Swarthmore/your undergraduate institution, make sure to ask your professors to write a letter of recommendation for you and keep it on hold BEFORE GRADUATING or NOT TOO FAR AFTER GRADUATION!! Profs will be able to write better letters for you when you are around and they remember you better.

Putting together your CV

Something that I found helpful was putting together my CV in Latex through Overleaf - here are some templates you can use!
You can and should put on relevant coursework that you’ve done at Swat (and elsewhere), along with a short bulleted explanation of what you’ve learned in the class!
- Also don’t forget to put down the textbook name and author if your courses had a textbook.
- This doesn’t have to be from just the direct field/area of research you’re going into - some other things I put down were Modern (Abstract) Algebra and Logic. Don't forget to put on the various work experiences you had outside of your intended research field/area! Sometimes having that well-roundedness could be what people are looking for.

On the topic of emailing potential advisors...

Please be careful about this section, as it really pertains to the experience I had. In areas that are not CS, maybe everything I said might be wrong. For example, I know that for Biology you would have to be in contact with the group that you intend to be a part of.
There are super contradicting opinions on whether you should or shouldn’t email professors of the schools that you are applying to. Sometimes people tell you that you must email potential advisors, and sometimes people tell you that you probably shouldn't email potential advisors.
To be honest, I think it doesn’t make a big difference whichever way you go. Let me explain…
- If you really love their research topic and read some paper(s) by the prof, and end up having specific questions about the content of the paper(s), then you should email them.
- Or, if you have any genuine questions about the field that noone at your home institution can answer, that would be very nice.
- Emailing professors usually won’t be the way you stand out from the other applicants. Most likely what would happen is that the prof gets so many emails that they either 1) don’t notice your email 2) don’t have time to read your email 3) even if they did, they might not even remember you/have the power to change the status of the app when application review season comes around.
- That being said, some profs do note on their website that they’d love to hear from prospective PhD students so… that might be good if you’d like to send them an email.
Conclusion: Emails are (in most cases) not the be-all end-all!! If you’re intending to read papers and send an email just for the sake of having a prof notice you, it probably isn’t worth the effort.

Interviews

Interviews can be really nervewracking!! (Speaking from personal experience) But in a lot of cases, the prof who is interviewing you is curious to know who you are, and whether you'd be a good fit to work with them!
Make sure to be yourself! Being genuine as possible will help the prof know you better.
Some things that are important to review are (1) the work that you've done so far and (2) the kind of things that you want to learn/the kinds of problems you'd like to tackle as a grad student (3) anything you found special about that program, and why you think the program you're applying to (or a prof that you want to work with) is cool!!
- Something that helps for me when I practice talking about my work is to have someone (preferably an advisor or a peer researcher) quiz me by simulating a conversation. It's actually surprising how difficult it is to cover all of the basis of what you've done in general terms, especially when you're so involved in a particular project!
- Show off your enthusiasm for this specific program!! And this is why it's very important for you to only apply to places that you'd be happy to go to :) The prof is looking for students who would be happy working with them!
Most likely the prof will ask you whether you have any questions for them. If you're doing a remote interview, what helped for me is to have a sheet of paper (or a word doc) with some questions and notes for the talking points I put above. That being said, make sure to not completely read off of these notes!
If the interview you are doing is a coding interview, it very well may be the case that the coding interview is assessing whether you have the skills to code with respect to your area of interest!
- For example, if you're interested in Programming Languages, make sure you can think and code in functional programs!
- That being said, take this with a grain of salt because I don't have too many experiences with coding interviews for grad school.