* The SOP is meant to a stream of text without sectional headings, but here in the interest of understanding, we are highlighting the various sections.

The first time I visualized a data set, I saw not just numbers but stories—patterns of behavior, gaps in access, potential futures. In my undergraduate capstone project, I developed a model to predict hospital readmissions using patient history and real-time vitals. The model’s predictive accuracy wasn’t what excited me—it was watching a hospital administrator use it to flag a high-risk patient and initiate timely care. That experience confirmed what I had long suspected: I wanted to build data-driven tools that improve lives. This conviction has led me to apply to the Master of Science in Data Science at the University of Toronto.

I graduated with distinction from Vellore Institute of Technology, majoring in Information Technology. While I was drawn early to programming and systems thinking, it was a third-year elective—Data Mining and Predictive Analytics—that sparked my interest in data science. Working with a team of peers, I developed a predictive model for dengue outbreaks in South India using climate and sanitation data. The project received accolades in our university tech fest and gave me a first taste of interdisciplinary, high-impact problem-solving.

Eager to apply these skills in the real world, I joined ZS Associates as a Data Analyst. Over two years, I worked with leading pharmaceutical clients to model drug adoption rates, analyze clinical trial metrics, and optimize salesforce allocation using tools like SQL, Python, and Tableau. One of my most rewarding projects involved building a churn prediction model for patient support programs, which helped increase retention by 12%. I collaborated closely with cross-functional teams and presented findings to stakeholders in North America, honing both my technical and communication skills.

To deepen my understanding of the field, I pursued the IBM Professional Data Science Certificate and completed multiple MOOCs, including Andrew Ng’s Machine Learning course and a specialization in applied data science with Python. I also contributed to an open-source COVID-19 dashboard that tracked hospital bed availability across Indian states, which received 10K+ views in its first week. These experiences have reinforced my belief that data science is not just a technical discipline—it is a tool for empathy, efficiency, and equity.

What draws me to the University of Toronto is not just its academic rigor but its emphasis on real-world impact. Courses like STA2453: Applied Machine Learning and CSC2541: Topics in Machine Learning directly align with my interests. I am particularly excited by the work of Professor Frank Rudzicz at the Vector Institute, whose research in AI for healthcare mirrors my own aspirations. I would also love to contribute to projects at the Schwartz Reisman Institute, especially those focused on ethical and socially-responsible AI.

Equally compelling is Toronto’s location in the heart of Canada’s data science ecosystem, providing exposure to industry giants, startups, and cutting-edge research hubs. I aim to leverage this ecosystem for internships and collaborations that bridge academia and practice. In the longer term, I plan to return to India and lead data science teams that develop solutions for public health optimization—be it in patient triage, epidemic forecasting, or medical supply chain logistics.

At U of T, I bring with me a unique blend of academic excellence, industry insight, and social consciousness. I have mentored three undergraduate women at ZS through our “She Codes” initiative, and intend to continue championing diversity in tech through clubs like WiDS Toronto and UofT Data Science Club. I believe that who builds the technology matters as much as what is built—and I am committed to shaping an inclusive, impactful future.

Data science, to me, is more than a career—it is a way to serve. With the University of Toronto’s MSDS program as my foundation, I hope to become a changemaker who uses data not just to answer questions, but to ask better ones—especially the ones that matter.