Writing a Personal Statement

As students, we often need to write personal statements of purpose for university admissions and fellowship applications. Here, I wanted to share my experiences writing personal statements over the years. I’ve previously written personal statements for fellowships and university admissions.

What helped me the most while writing these statements was actually seeing other people’s examples, so I’ve provided my statements from the past and links to websites that contain many examples:

I’ll describe the approach I found logical for writing a personal statement. This may not be the best way for everyone, but it worked for me.

Audience

Understanding the purpose of your writing is crucial, which means asking yourself: “What do I want my audience to get out of this?” For personal statements, this varies based on the audience. Depending on who they are, I emphasized different experiences to match the desired outcome.

In fellowships, they often explicitly state what they’re looking for in a solicitation document. If none is available, contacting the fellowship officer for clarification is an option. For university admissions, you can reach out to departments to understand their selection criteria.

Structure

The structure I used for my NSF Graduate Fellowship personal statement was:

Introduction
Intellectual Merit (Experiences)
Broader Impacts
Future Goals

For fellowships, a recommended structure is often provided, so refer to specific fellowship guidelines. Generally, the structure is designed to present a cohesive story that addresses all essential points.

Introduction

The introduction should draw the reader in, and many applicants do this through a personal anecdote. Here’s an example of the approach I used for the NSF fellowship:

“python inputBankingInfo.py Running… Finished. And with that, it was done.” The perpetual twelve hours of weekly data entry into Excel spreadsheets had forever ended. As an assistant real estate manager, I was tasked with repetitive work, like data entry. Every day, I slogged through inputting tenants’ accounts payable into an Excel sheet, thinking there had to be a better way. That’s when I discovered Automate the Boring Stuff with Python, my first programming book, and automated the data entry process—marking the beginning of my journey as a programmer.

Many successful introductions start with a story that engages the reader. For me, this example outlines how I got into programming. I’ve found that writing an introduction in a style you’re comfortable with is important; avoid forcing a tone that feels unnatural.

Next, I used a time jump to bring my story to the present:

Four short years later, I have written code for Amazon, one of the largest companies on the planet; taught hundreds of students about machine learning (ML) and artificial intelligence (AI); participated in hackathons across the country; researched synthetic human-like speech generation; now I’m working towards creating robots that can execute tasks defined in natural language, while also exploring ways to communicate with computers directly through our brains.

I then addressed a potential weakness in my application:

My journey into research wasn’t straightforward. I didn’t even know what it meant to do research until the summer before my third year of university, and I didn’t stick to a single lab throughout my undergraduate years. Although this may have resulted in fewer publications, it helped me discover my interests, become more independent as a scientist, and strengthened my conviction that I want to pursue a career in research.

Here, I acknowledged my limited publication record—a common metric of research success—while framing my diverse experiences positively.

Experiences

The experiences section is usually simpler to write, and I found the STAR method effective. Commonly used in interviews, it helps in writing about experiences concisely:

Situation: Provide background.
Task: Explain what you were trying to accomplish.
Action: Describe the steps you took.
Result: Highlight the outcome.

Here’s an example from my statement:

Situation:

During the summer before my junior year, I collaborated with Dr. Kangkook Jee in his System Security Lab at the University of Texas at Dallas (UTD).
Task:

My role was to extend the Graph Neural Network (GNN) Explainer algorithm to work with heterogeneous graphs, enabling us to generate explanations about why the GNN classifier detected a program as malicious based on the corresponding heterogeneous resource interaction graph.
Action:

After programming the algorithm, I validated its accuracy by applying it to a classifier trained on the MUTAG (Mutagenic Chemicals) dataset. I confirmed the reliability of the algorithm’s explanations by comparing the molecules it identified as most important for classification as a mutagenic compound to established scientific ground truths, finding a strong match with existing research.
Result:

This work led to my open-source contribution of the algorithm to the Deep Graph Library, the second-largest GNN library, and the algorithm was later used in the lab’s research publication.

Sometimes, even with a strong structure, an experience may not fit well into your narrative. In such cases, it’s worth pursuing new experiences that better support your story.

Writing Results

When writing about results, it’s crucial to articulate the impact of your actions clearly. This can include tangible outcomes, like speed improvements, cost savings, or research publications. But when tangible metrics aren’t available, you can still highlight what you learned or the skills you developed.

For example, a common job experience, like being a cashier, can be written from the perspective of the impact you made. Instead of saying, “I worked as a cashier at Walmart for four months,” you could frame it as: “I helped process approximately $100,000 in transactions over a four-month period, ensuring smooth and accurate customer interactions.” This shifts the focus to the benefit you created for the employer, making your contribution clearer and more impactful.

Being creative with how you present your results can make a big difference. However, it’s essential to stay truthful and not exaggerate your impact. The goal is to frame your achievements in a way that emphasizes your contributions without distorting reality.

Broader Impacts

Most personal statements require a section demonstrating how you’ll benefit others or society. For the NSF, this involves broader impacts like broadening participation in STEM. This section follows a modified STAR approach, with an added focus on future plans (STARF):

Situation:

When I first began learning about machine learning, I found the mathematical concepts intimidating.
Task:

This inspired me to help others by explaining these ideas in a more accessible way, only requiring a basic understanding of mathematics (such as understanding the slope of a line).
Action:

To that end, I became the workshop lead for the Artificial Intelligence Society, a campus organization, where I developed and presented eleven workshops and led a team of four to create nine of those workshops. I particularly encouraged students from various backgrounds to attend my workshops, even if they didn’t have a background in computer science. These workshops covered areas such as Deep Learning, Computer Vision, Natural Language Processing, Model Deployment, and Reinforcement Learning.
Results:

The largest event had around 300 participants, while on average, each workshop attracted 57 students. Our feedback data showed an average engagement rating of 4.1/5.0, with 96% of participants indicating they learned something new.
Future:

Moving forward, I plan to continue educating students through workshops and YouTube videos, presenting AI techniques in a simple and accessible manner to foster a diverse and vibrant AI community.

This approach links past impacts to future plans, making them credible.

Future Goals

The future goals section ties everything together. It should connect your introduction, experiences, and broader impacts, showing how the opportunity you’re applying for will help you achieve your objectives and contribute to the organization’s goals.

Objective:

After completing my master’s, I plan to continue my research in robotics by pursuing a PhD in Computer Science and later becoming a research scientist. In my research career, my primary goal is to develop autonomous robots capable of reliably performing a wide range of tasks defined by humans, using natural inputs like text.
Experiences and Impacts:

My experiences in both research and industry have equipped me with the technical and interpersonal skills necessary for a successful research career. Additionally, my experiences running workshops and helping students find jobs have instilled in me a deep commitment to educating and assisting others.
Support and Alignment:

Receiving the NSF Graduate Research Fellowship would enhance my future research by giving me the flexibility to explore areas I find both interesting and impactful, as well as the freedom to continue to help other students. Awarding the fellowship to me would enable the NSF to further its mission of advancing artificial intelligence in robotics and broadening participation in the tech field.

Writing Process

My writing process generally follows four steps:

Outline: I outline the structure and organize the story.
Dump the details: I dump all details into the sections of my outline.
Edit: I revise for readability and cohesion. This involves rereading sections or using AI tools for assistance—whatever works best.
Review: I have others review my writing for readability. Incorporating their feedback is important to ensure the writing flows smoothly and is easy to understand.

If you have further questions, feel free to reach out to me.