Do you find it difficult to “sell yourself”? Most CS students do, at least in my experience. It’s a funny quirk. When you ask what they want to study in the future, too many respond with “Artificial Intelligence” or “Machine Learning”…and then they say no more. Very few respond with specific ideas about the problems they want to solve. Yet, when thinking about how to write a computer science statement of purpose, specific problems are everything.
We’re not selling ourselves. We’re selling our ideas. We’re selling the potential of our minds.
Just ask Ray, a specialist in Human-Computer Interaction who received a whole slew of admissions last year. After using the Structure is Magic template, Ray’s SOP was so incisive that it prompted one professor at his top school to write:
“Wonderfully written, ambitious, and aligns perfectly with the work we do at our lab.”
Can’t ask for a better response than that, can we?
Luckily for you, Ray’s essay offers lessons you can extract and use right now. He followed Structure is Magic closely, crafting a timeless and elegant argument. He expressed big and compelling ideas. He showed a concern for the greater impact of his work on society, and he described exactly how the professors at his target school can help him make that impact.
Want to write a computer science statement of purpose that earns multiple fantastic admissions? Read on and learn how.
Before we read the essay itself, let’s look at how it’s structured into 4 neat sections. Each of the first 3 sections convey, in order, answers to the questions below. Then, the final section reaffirms them.
- Exactly which hyper-focused academic questions are most important to me, and how did I discover them?
- How will this graduate program help me solve these problems?
- What convincing proof do I have that I’m 100% ready to pursue these questions as a graduate student?
Thus, we can break Ray’s essay down as follows:
- Introductory Frame Narrative – The quick story of how Ray discovered the intellectual problems he wants to solve in grad school and then in his career.
- 2 paragraphs
- 28% of word total
- Why This Program – How this specific university will help Ray solve the problems he mentioned in the previous section.
- 2 paragraphs
- 27% of word total
- Why You’re Qualified – The academic and career highlights that prove Ray is capable of actually succeeding as a graduate student pursuing these problems.
- 3 paragraphs
- 36% of word total (this is where the essay gets too long)
- Closing Frame Narrative – Quick flashback to the intro, and reaffirmation of goals.
- 1 paragraph
- 9% of word total
Sections 1 and 2 (the Introduction and “Why I’m Qualified”) are probably too long, but other than that, Ray’s SOP is structured perfectly. For other students, I’d suggest they shorten sections 1 and 3 to less than 25% of the word total each. Maybe use that extra space in Section 2 and Section 4. Or…don’t add anything else at all. Remember: master’s applicants usually shouldn’t write more than 850 words.
In my experience, CS students, more than anyone else, love to write long “Why I’m Qualified” sections. They love to list hackathons they’ve won, internships they’ve held, classes they’ve taken, and programming languages they’ve learned. But it’s important to remember that this section isn’t your CV. It’s like a small box that only lists the highlights from your CV – the most important parts. Certainly not everything. In Ray’s case, his experiences were so unique that it was hard to cut them down. But you should. Cut them down as much as you possibly can.
Why? Because that’s not the most important part of the essay. The academic ideas you express, and your fit with the program – that’s where you truly sell yourself.
Wonderfully Written. Ambitious. Aligns Perfectly.
What’s so great about Ray’s SOP?
As you’ll see shortly, this essay tackles huge topics. It asks big questions. It shows a tremendous human desire to answer those questions. And, in its own engineer’s kind of way, it shows concern for real people.
Most CS applicants don’t dive very deeply into their “study plan” for grad school — they just list classes they want to take, and maybe mention a professor or two in passing. But Ray speaks like someone who’s worked in HCI and AR for years. He knows exactly where he wants to probe further, and knows exactly which professors are the best guides on this journey.
(Ray, it should be obvious, did plenty of research on his target programs.)
As you read, pay close attention to the questions Ray asks. He’s asking them to the professors at his target school. At the same time, he’s asking them to himself. When an answer appears, he tears it apart and finds the hidden questions inside. This is way deeper than most MS applicants go. Heck, it’s deeper than many PhD applicants go!
Through this process, Ray shows that he’s more than just a guy who can plug some Python into IDLE or iterate a Map in Java. He’s a thinker, a creator, an innovator. Then, when he explains how three specific professors at his dream school are already working on these questions that he’s just asked….well, that’s what we call a coup de grâce.
- He asks 6 questions.
- He describes how 3 professors are currently tackling those questions.
- He shows how he’s spent years preparing to work on these questions.
What do you think, should this guy get admitted? Let’s read his SOP and decide.
A Rockstar Computer Science Statement of Purpose
During my 2019 internship at JL Design Partners, an architecture firm in Sacramento, I was tasked with developing an Augmented Reality application for visualizing building designs. At that point, I was already intrigued by AR’s possibilities in diverse fields, from entertainment to aiding surgeons. Then, after two weeks of writing C# in Unity and establishing a workflow from Autodesk Revit, I loaded the prototype onto my phone and ecstatically invited my colleagues to try. Immediately, however, as I watched them rotate the virtual 3D structure by swiping on a 2D screen, I understood (with dismay) how counterintuitive the interaction was. At the time, I realized that products like daily-wear AR glasses could make architectural visualization more intuitive, but due to challenges with optics, batteries, and spatial understanding, these do not yet exist. Now, I wonder if solving these problems will be enough for widespread adoption at all. To what extent are we clinging to formats that do not present an effective way to interact with virtual content?
Which type of interface most complements AR? Which most improves cognitive processing and decision-making? Brain-Computer Interfaces are the common answer, but if so, what is the best approach to combining AR and BCI? What kind of neural activity should be employed? How can we design AR experiences for reactive BCIs? These questions lingered in my mind in the months that followed my internship. They remained there well after earning my degree in Computer Science, and even today, they resonate with the very same frequency. This is precisely why I seek admission to the Master’s Program in Computer Science at Stark University: to ascertain the optimal approach for virtual interactions that reduce cognitive load, enrich the senses, and augment human capabilities.
Perhaps no other graduate program in the world is more uniquely suited to helping me explore this niche within Human-Computer Interaction. The variety of interdisciplinary research carried out at the Stark HCI Lab, and the fact that it is one of the few funded master’s program that allow specialization in HCI, all compel my decision to apply. Certainly, coursework such as MCS731 Constructing Cross Reality Applications and MCS728 Affective Computing will prove essential as I begin to comprehend the core principles and design guidelines required for a thesis studying responsive and instinctual interactivity. Most importantly, however, is the opportunity to conduct research under Dr. Bruce Banner, Dr. Erik Selvig, and Dr. Jane Foster, whose work aligns perfectly with my goals of enhancing human cognition.
Particularly intriguing is Dr. Banner’s study on Electrodermal Activity and its relationship to physiological arousal. I hope to explore how this can be implemented as a metric to uncover cognitive loads of atypical input modalities. I am also captivated by Dr. Selvig’s work in augmenting user perception of the environment, and hope to investigate how eye-tracking can strengthen the system’s perception of user intent in HearThere, and how the sensor network in Tesseract can be used for augmented on-site exploration through persistent anchors. Additionally, I am open to working under Dr. Foster as her projects on musical expression and therapy (Vocal Waves) are similar to my ongoing work, Fluid Fingers. Having worked on numerous similar projects, I am confident I can make substantial contributions to these and other projects at Stark, while simultaneously exploring esoteric approaches like functional near-infrared spectroscopy for its superior spatial resolution (as compared to EEG) and resistance towards motion artifacts.
Perfect as the opportunity may seem, I am well aware that the daunting problems of AR require far more than general competence. Fortunately, my academic and research experiences have equipped me with the tools to face them. I was introduced to immersive technology in undergraduate coursework at Empire State University, such as Computer Graphics and Virtual and Augmented Reality. The Interaction Design Specialization from UC San Diego (via Coursera) further improved my understanding of Ideation, Prototyping, and Evaluation methods. Additionally, during my final undergraduate year, I developed a framework, ILMR, that allowed for prototyping hand interactions through direct manipulation with virtual content. Inspired by Google Cardboard, I conceived it using inexpensive and accessible modules like the Zeiss OnePlus headset and the Leap Motion Controller. However, it lacked depth perception and haptic feedback, which are vital in letting the user know they have successfully interacted with virtual content. Through a collaboration with Dr. Otto Octavius, a postdoctoral researcher at the Osborn Private Science Institute, we developed an interface, OptiTouch, that provided the semblance of haptic feedback using only visual cues.
My most rewarding experience, however, was not with ILMR, but rather the research I conducted unofficially with Professor Charles Xavier after graduation. I was intrigued by Neurofeedback and its non-pharmacological approach towards mental health therapy, but felt the feedback modality lacked appeal. Keeping ADHD-diagnosed children in mind, we devised an AR telekinetic experience where a user could bend a virtual spoon by reaching the desired psychological state. This work was accepted to the International Symposium on Mixed and Augmented Reality, where AR academia veteran Dr. Reed Richards claimed it to be impressive.
In my free time, I actively participate in hackathons, where I have led teams to multiple triumphs and developed multiple AR apps. One such was a music visualizer, VizAR, now published on the App Store. I redesigned the particle system in VizAR for Head-Mounted Displays, and enabled hand-particle interactions by implementing a Signed Distance Field. This ultimately led to my work, Fluid Fingers, which aims to elicit emotional states akin to those exhibited in live music performances. Considering these experiences, I feel confident that I am prepared to hone my abilities even further at the Stark Lab.
During my architectural design internship two years ago, I saw immediately the great chasm between our current technological capabilities with AR, and normal users’ ability to employ them. The questions that arose have been my guiding force ever since. This is precisely why I seek to contribute to the field of Human-Computer Interaction, and eventually work as a researcher in industry, ensuring AR’s impact in people’s everyday lives. To begin this effort in the Stark MS CS program would be the most tremendous honor of my career.
What should you take away from this essay?
- You should absolutely mimic the structure. Unless a specific school requires otherwise, there are no more effective ways to write a general statement of purpose.
- You should absolutely state, with 100% clarity, the questions you want to study in grad school.
- You should absolutely describe how professors in your target program are already studying these questions, and what you’d like to investigate under their guidance. If you don’t know already, then spend some time on the department website, or use a resource like Drafty to find professors who match your interests.
I’m indebted to Ray for allowing me to republish his SOP and to brag about his success. No matter which CS field you want to specialize in – whether HCC, Network Engineering, Computational Linguistics, or something else – there’s a lot you can learn from Ray’s example.
Ask big questions. Show that you have a real, thoughtful purpose. Show that you want to make a difference in the world. Ultimately, that’s what it’s all about, right?
Then, if you can link those questions to specific resources, labs, or professors at your target program, you too will have a rockstar SOP on your hands.
“Wonderfully written…ambitious…aligns perfectly.”
You can get the same feedback, friend, and now you know how.