I'm a Machine Learning Researcher at Apple, where I work on post-training large language models for alignment, reasoning, and open-domain question-answering.
My work centers on post-training Large Language Models for Apple Intelligence, specifically Supervised Fine-tuning, Distillation, Preference Optimization and Reinforcement Learning to improve reasoning, multi-turn and tool use capabilities and making responses grounded, factual, trustworthy, and generalizable.
Before Apple, I was a Research Scientist at IBM Research in the Brain Inspired Computing Group, where I developed novel algorithms for training low-precision neural networks for efficient inference at the frontier of energy, space, and time, and contributed to the development of Northpole, an inference accelerator published in Science, described in this Nature article as a "Mind-blowing AI chip".
When I'm not doing research, I love being outdoors — city walks, nature hikes, bike rides, or discovering new matcha and coffee spots.