About
I am a Research Fellow at Microsoft Research under Nagarajan Natarajan working at the intersection of LLM post-training, large-scale distributed training, and low-level systems performance. My research spans code generation models, parameter-efficient fine-tuning, and high-performance inference. I am also a core maintainer at NumPy org.
My current work includes autoformalization and verification pipelines for Rust and FP8 inference kernels for large language models.