Pankti Sheta

I started as a Research Fellow where I got to work on the Nest Fact Checking Foundation Project, where we analyzed Facebook content from Chinese and Russian state-linked pages using keyword datasets. I helped surface top content, track external links shared across posts, and generate data visualizations that are actually useful to non-technical users. This project gave me a strong foundation in exploratory data analysis, and working with real-world, messy social media data—skills I can now use confidently.

When I came back as a Research Engineer Intern, I got to roll up my sleeves and work on Arbiter, one of SimPPL’s key tools. My primary focus was building a robust topic modeling pipeline that could extract key themes and evolving narratives from large-scale social media data. I worked on developing and refining Named Entity Recognition (NER) components enabling the extraction of key entities from user-generated text for downstream analysis. This made me realize, my code was no longer just sitting in a GitHub repo—it was part of something people actually used. That was huge for me. I’d put in so much effort in learning how to build things, but this was the first time I truly felt like I was helping someone with my skills.

These technical and soft skills played a huge role in helping me get accepted into the University of Maryland for my Master’s in Data Science. When I wrote my Statement of Purpose, so much of it reflected what I had learned at SimPPL—about the real-world relevance of data science, the ethics of technology, and the importance of working on problems that actually matter.

But the best part? The people. SimPPL wasn’t one of those places where you’re just a cog in a machine. Dhara and Swapneel didn’t just give me tasks—they gave me room to learn, ask questions, and even mess up sometimes. They treated me like someone who could contribute, not just an intern or a newbie. That kind of trust changes how you see yourself.