I am a second-year graduate student at Carnegie Mellon University, pursuing my Master's in Machine Learning.
My interests lie broadly in the science of foundation model training, understanding RL, and diffusion over sequences. Most recently, I was a Research Scientist Intern at Mistral AI, where I worked on the end-to-end RL post-training of code agents, from training environments and reward design to evals.
At CMU, I have also worked with Prof. Max Simchowitz on understanding gradient dynamics during multi-objective reinforcement learning for robotic policies. Our proposed drop-in intervention mitigates pathologies associated with gradient interference.
Previously, I graduated from the Indian Institute of Technology Bombay, where I majored in Computer Science and Engineering with honors. I have been fortunate to be advised by Prof. Devendra Singh Dhami at TU Eindhoven and Hessian Center for AI on my research around probabilistic circuits and optimization (NeurIPS '25, UAI '24). At IIT Bombay, I had the privilege of working with Prof. Preethi Jyothi for my bachelor's thesis. We researched flow matching and other generative approaches to compute invariant representations of speech, to improve recognition performance for non-native speakers.
In my spare time, I love to play all kinds of sports: football, basketball, badminton, and table tennis being my favorites. I also love to write, and I sketch sometimes.
A novel method for predicting sparse granger causal relations between a set of time series variables, using an xLSTM based architecture and a dynamic lasso penalty for inducing sparsity. Sepp Hochreiter recognised our work!
We compute post-intervention likelihoods based on true causal relationships between variables. We developed a novel tractable probabilistic circuit, that enabled causal inference for hybrid domains, that is, on systems with both discrete and continuous random variables. By using characteristic functions to represent distributions, we were able to better model asymmetric and long-tailed distributions, which are especially prevalent in real-world datasets.
