About
I am a Senior Research Software Engineer at Google Research NY. My work is focused on foundational research in machine learning and how these theoretical insights can be applied to improve ML efficiency, especially in the context of fine-tuning.
Previously, at Google, I was an ML Solutions Engineer in Google Cloud working closely with Cloud customers to build and productionize machine learning models at scale. In my role within the Advanced Solutions Lab, I also led a 4-week intensive training course teaching the tools, techniques, and best practices for training and deploying their own ML solutions.
Prior to Google and my switch to industry, I had a career in academia as a research mathematics professor in the field of geometric analysis.
You can find my CV here.
Research
I am broadly interested in the theoretical foundations of machine learning and how to leverage those insights to devise more efficient training and optimization algorithms. Recently, I have been thinking a lot about transfer learning and fine-tuning, particularly how implicit biases can be leveraged to improve the training dynamics of large deep learning models.
You can find a list of my research work here.
Books
I’m also a published O’Reilly author and have written two books on machine learning. These are not research books but instead intended for a general audience of ML/AI Engineers and practitioners. You can see short summaries here.
Community
In addition to my primary role at Google, I also volunteer as an AI Coach with Google.org, the philanthropic arm of Google. I advise and mentor early and late-stage startups using AI for social good. I’m currently participating in Google.org’s Generative AI Accelerator program. You can find description of this and previous volunteer work with Google.org here.