Michael is currently leading several research groups focusing on using large language models for retrieval augmentation, ranking and query/document understanding. Michael is broadly interested in practical applications at the intersection of information retrieval, natural language processing, and machine learning. Specifically, he has worked on research problems in a variety of domains, including search (web. social media, news, email & enterprise), recommendation systems, document clustering, web crawling, query intent classification, information extraction, plagiarism detection, e-commerce and search advertising.