I enjoy building and deploying algorithms and ML models with mathematical rigor for scientific, engineering, and technological applications.

Currently at The New York Times, I develop and deploy real-time ML models for subscription-related problems, such as serving a paywall at optimal moments or personalizing messages to drive subscriptions. Models I build are often constrained and multi-objective in nature, sitting atop causal ML and reinforcement learning algorithms such as contextual bandits.

I write statically-typed production code in Python and Go to support cloud-based containerized applications for high throughput inference. I have also built production-scale batch models that determined user-level access to Times content (see this blog post, this PyData talk, or this media article by VentureBeat). See my CV for more details.

At the Times, I have applied causal inference to observational data using statistical matching techniques. I am also involved in designing and launching Randomized Control Trials (RCTs) to collect data for training causal ML models. During an internship at Amazon in 2020, I built a Double (De-Biased) machine learning model to estimate the causal impact of Advertising products.

My Ph.D. thesis at MIT was advised by Prof. Jörn Dunkel in the Applied Mathematics department. My work combined ideas from scientific ML, mathematical modeling of active matter, and numerical computation. I built a computational inference framework to learn continuum models from microscopic data to predict collective phenomena in systems such as bacterial suspensions or self-propelled colloidal particles. I was a recipient of the Mathworks Engineering Fellowship. I also received a Master’s degree from MIT for my work on theoretical and experimental modeling of oceanic internal waves. During my time at MIT, I enjoyed teaching and was a teaching assistant for Advanced Fluid Mechanics (1 semester) and Dynamics and Controls (4 semesters).

In 2015, I completed my B.Tech in Mechanical Engineering at IIT Madras, India. I worked with Prof. Mahesh Panchagnula on theoretical fluid dynamics research and spent a year building an omni-directional mobile robot for the ABU Robocon competition to solve pick-and-place and throw tasks (that’s me driving the robot!).

Outside of work, I enjoy long-distance running in the summer and skiing in the winter.

Skills

Machine Learning

Supervised and unsupervised machine learning, Contextual bandits, Bayesian methods (Thompson Sampling), Causal Inference, Causal Machine Learning, Neural Differential Equations, Scientific Machine Learning

Mathematical Modeling

Differential Equations, (Fluid) Dynamical Systems, Optimization, Linear Algebra, Numerical Methods for PDEs

Technologies

Programming

Machine Learning

ML Deployment