I graduated from Rutgers with a B.S. in Computer Science and Data Science, twice on the Dean's List, and most of my work since has been about making complicated technical material actually make sense to someone who didn't build it. I co-authored research on graph neural networks that predict aerospace component failures, and a core part of that work was explainability: figuring out how to explain why a model made a prediction in plain terms, not just getting the prediction right. I later...
I graduated from Rutgers with a B.S. in Computer Science and Data Science, twice on the Dean's List, and most of my work since has been about making complicated technical material actually make sense to someone who didn't build it. I co-authored research on graph neural networks that predict aerospace component failures, and a core part of that work was explainability: figuring out how to explain why a model made a prediction in plain terms, not just getting the prediction right. I later founded an AI systems project where I ran dozens of interviews with financial analysts, translating their vague, half formed problems into concrete requirements. Whether it's a model's reasoning or a person's confusion, the job is the same: find the actual gap in understanding and close it.
My tutoring experience is informal but real. I've tutored family and friends in calculus one on one, and my approach is to build intuition for what a derivative or integral actually represents before touching the formula, since most confusion comes from memorizing rules without understanding what they mean. In college I regularly helped classmates in data structures and physics study groups, working through problems together instead of handing over answers, and having them explain a step back to me so I could catch exactly where their reasoning broke down. I also spent a software internship working alongside less experienced interns, walking them through unfamiliar code at whatever pace actually worked for them.
My calculus tutoring has mostly been one on one with high school and early college age students. The data structures and physics support was in small groups with college peers. Across all of it my approach stays the same: meet someone exactly where they are, figure out precisely what's unclear, and explain it in a way that sticks.
