I am a 4th year PhD student at Stanford developing new machine learning methods, particularly those with applications to medicine and biology.

I have developed new methods to analyze genomics data, ultrasound imaging, time series data, and other kinds of biomedical datasets that have been published in machine learning conference (ICML, NeurIPS) as well as peer-reviewed journals (Nature Communication, Nature Genomics, Nature Digital Medicine and Nature Machine Intelligence).

I have been fortunate to work with several companies to help deploy machine learning systems. Here are some examples of work that I have previously done:

• Fusing data modalities to detect falls: I worked with a company that builds smart canes. My role was to implement algorithms to detect when a user has fallen. By leveraging different data modalities, we were able to build a robust detection system.
• Deep learning with missing data: I consulted remotely for a Fortune 500 company to perform deep learning on time series data. The challenge: much of their data consisted of missing values. By leveraging intelligent imputation methods, we demonstrated a significant improvement over classical machine learning algorithms.
• Summarizing clinical notes with little data: I worked with an early-stage startup that was building systems to summarize clinical notes. The challenge was that they had very little data. By leveraging pretrained word embeddings and heavy data augmentation, we were able to help them build a summarization system to share with early customers, providing a significant ROI.

Contact me

If you’d like to work with me, or schedule an initial consultation, shoot me a quick email at: a12d@stanford.edu.