About me
I’m Arkajyoti (AJ) Bhattacharjee — a Ph.D. student in Statistics at The Ohio State University, advised by Prof. Arnab Auddy, and pursuing a graduate minor in Computer Science. My research focuses on Differential Privacy and Machine Learning: designing algorithms that balance privacy with utility, proving guarantees, and validating them through large-scale experiments. I design methods that don’t just look good on paper but run, reproduce, and hold up under pressure.
My path here began at Presidency University, Kolkata, where I first discovered how probability and inference could cut through complexity. I went on to the Indian Institute of Technology, Kanpur for an M.Sc. in Statistics. There, with Prof. Dootika Vats, I worked on variance estimation for adaptive Markov Chain Monte Carlo. That project forced me to reconcile theory with stubborn simulations, teaching me a rule I still live by: theory matters only if computation confirms it.
Alongside academic research, I’ve pursued projects that make statistical work reproducible and impactful. During Google Summer of Code 2021, I contributed to the R ecosystem by improving nonlinear least-squares tooling. At Accenture Solutions, I built an AutoML prototype that streamlined preprocessing, model selection, and visualization. At OSU’s Statistical Consulting Service, I’ve worked with researchers in speech & hearing science, geography, rehabilitation, and anthropology, translating complex methods into tools and insights they could rely on.
Teaching is another constant in my journey. As a Graduate Teaching Associate, I run labs and recitations, guiding students as they turn abstract equations into working analyses. For me, clarity comes when mathematics, computation, and intuition all line up — and helping students reach that point is one of the most rewarding parts of my work.
Outside the academic grind, I train consistently, play racket sports, and enjoy disc golf. The discipline, progress, and structure I practice in the gym mirror how I approach research: steady effort, reproducibility, and the drive to get stronger with every iteration.
In short, I’m building a career at the intersection of rigorous theory, scalable computation, and reproducible practice. If you’re looking for work that holds up under stress — whether in theory, code, or application — we should connect.
Research interests
- Differential privacy
- Statistical learning
- Nonparametric statistics
- Markov Chain Monte Carlo
- Stochastic optimization, Bayesian/variational inference
- Reproducible software & scalable experiments