소개
I am the Principal Architect leading software engineering efforts on the Immunomics (opens in new tab) team in the Healthcare Research (opens in new tab) group at Microsoft. In partnership with Adaptive Biotechnologies (opens in new tab), our goal is to create a universal blood test that reads a person’s immune system, to detect a wide variety of diseases including infections, cancers, and autoimmune disorders in their earliest stage, when they can be most effectively diagnosed and treated. The basis of this project is learning the T-cell Antigen Map (opens in new tab): that is, learning the mapping from a T-cell receptor sequence to the antigen(s) it targets.
My current efforts center around building systems to enable rapid and reliable machine learning exploration among our core team, key contributing researchers, and various partners. Generally, I aim to abstract the mechanics of data manipulation, code quality, and compute management to allow machine learning experts to focus on innovation.
Prior to joining Microsoft Research, I lead a team at Spotify building tools to help product owners and data scientists more quickly generate novel hypotheses for user behavior and construct causality experiments. This work was an extension of techniques developed at Preact, a technology company that I co-founded, which was acquired by Spotify for internal use.