
Blog
Alumni Profiles Series: Anh Do
Dr. Trang Anh Do is a data scientist at Google with a Ph.D. in Political Science from Duke University. Originally from Vietnam, she moved to the United States during her final year of high school and later attended the University of St. Thomas in St. Paul, Minnesota. After initially planning to pursue academia, she began exploring alternative career paths through conversations with alumni at Duke and later joined the Insight Data Science fellowship. She started her industry career at Nielsen in Chicago, moved to Meta in the Bay Area, and eventually joined Google, where she works on large-scale data and experimentation systems. Her career reflects the value of doctoral training in solving complex problems, managing ambiguity, and adapting to new professional opportunities.
How would you describe your journey, both professionally and personally?
I grew up in Vietnam and came to the United States during my last year of high school, which was a major cultural and academic transition. After college, I went on to pursue a Ph.D. in political economy at Duke.
At the time, I assumed I would stay in academia, since that was the most common path in my program. A key shift came during graduate school, when I joined the Emerging Leaders Institute. Through that program, I started interviewing alumni about their careers. Those conversations were the first time I seriously began to see alternatives to academia.
How did you go about finding an industry job?
Honestly, it was a process of trial and error. I spoke with alumni across consulting, think tanks, nonprofits, and data science roles, and I tried not to narrow my focus too quickly in any single direction. I was also transparent with my advisor and committee about the fact that I was exploring industry. That made the process smoother and allowed me to stay fully engaged in my academic work while preparing for the transition.
That curiosity eventually led me to Insight Data Science, an intensive boot camp designed to help Ph.D. students transition into industry by translating research skills into applied data science work and preparing them for the job market. Earlier conversations had been more exploratory, but Insight helped structure my job search more concretely by translating Ph.D. skills into industry-style projects and preparing me for interviews. That made the transition feel much more actionable.
From there, I joined Nielsen as a data scientist in Chicago, then moved to Meta in the Bay Area, and more recently to Google/YouTube. Each step exposed me to data problems at a different scale, from applied causal inference to large-scale experimentation systems.
Were there any major turning points in your career path?
There were a couple of key moments. The first was joining Nielsen. That was my first real exposure to industry, and I realized that I genuinely enjoyed applied work, especially using causal inference and data to answer practical business questions. Many of the skills from my Ph.D., such as structuring long-term projects and working through ambiguity, translated surprisingly well.
The second turning point was moving from Nielsen to big tech. The scale and pace were very different. Nielsen was a smaller, more contained environment and a great place to start. Over time, I wanted to challenge myself in a larger system.
I actually did not get into Meta on my first attempt. A few months later, after further preparation, I reapplied and was successful. That experience reinforced for me that career transitions are often iterative rather than linear.

How has AI impacted your industry?
AI has changed my day-to-day workflow quite a bit. I use it regularly for coding, debugging, and even drafting or refining analyses and presentations. One major shift is that data science is becoming less about writing every line of code manually and more about guiding, validating, and critically evaluating AI-generated output. Because of that, statistical intuition and domain understanding have become even more important. Without them, it is easy to over-trust outputs that may look correct but are not actually appropriate for the problem.
Any advice for current Ph.D. students?
One thing I often see is that people underestimate how transferable their Ph.D. skills are. A Ph.D. trains you to break down complex problems, learn independently, and communicate clearly—skills that are highly valuable in industry. I would encourage students to talk to alumni early and explore broadly rather than narrowing their focus too quickly. Those conversations were very helpful for me.
On the technical side, I would focus on statistics and Python. Tools evolve quickly, but the underlying thinking and problem-solving skills stay consistent.
If there were one thing you could go back and tell yourself during your school years, what would it be?
I would tell myself not to get discouraged by how long and uncertain the Ph.D. and job-search processes can be. There are many moments when things do not go as planned. I would also remind myself not to assume that there is only one “correct” path. Careers often unfold in ways that are not predictable.
And finally, I would say: do not be afraid of rejection. It is part of the process, not a final verdict.
AUTHOR

Chenyu Zhang
Ph.D. student, Political Science
Chenyu Zhang is a Ph.D. student in political economy at Duke University and a graduate of Williams College. His research interests include international trade, production networks, economic development, and social polarization, with a particular focus on how institutions and economic structures shape political and social outcomes. He enjoys skiing, playing the piano, and playing tennis outside of academics.