Lagunita Theme

Biography

I am a Research Scientist at Microsoft working on training and evaluating large language models for software engineering in the AI for Code team.

Previously, I worked as an Applied Scientist at Microsoft, focused on advancing reliability, reasoning, and efficiency in large language models. My work spanned confidence estimation and hallucination detection in LLMs, post-training and fine-tuning of LLMs, and designing agentic workflows with tool use.

Prior to joining Microsoft, I earned my PhD in Electrical Engineering from Stanford University in January 2024 as an NSF Graduate Research Fellow, advised by Prof. John Pauly and Prof. Mert Pilanci. During my PhD, I worked on diffusion models, probabilistic reasoning with LLMs, computer vision, memory-efficient learning, self-supervised learning, robustness under distribution shifts, and medical/computational imaging.

Earlier, I held research internships at NVIDIA (Learning and Perception Research Team), Microsoft Research, and ETH Zurich, working on robust diffusion models, probabilistic inference for LLMs, and ultrasound imaging, respectively.

I am passionate about research with meaningful product impact, bridging cutting-edge AI advances with real-world applications. Outside of work, you can find me on the tennis court, playing basketball or soccer, hiking, or perfecting my pool game.