I am a Senior Researcher/Software Engineer on the People + AI Research (PAIR) team in Google Research. My work focuses on interpretability and model understanding for large language and multimodal models, including feature and example-level attribution methods, counterfactual analysis, and concepts in embedding spaces. I am based in Boston, but collaborate with researchers across many locations.
I completed my PhD in Boston University under the supervision of Prof. Venkatesh Saligrama. My main focus was machine learning with natural trade-offs such as cost/quality trade-off at inference in computer vision models, and bias/representation trade-off in language models.
No news so far...