I have broad interests in applying Bayesian statistical modeling to web applications. My Ph.D research focuses on graphical models and model selection for high-dimensional problems. While at Yahoo! Labs, I have worked on several different web application areas, including recommender systems, content optimization, web fraud detection, computational advertising and so forth. After I moved to LinkedIn in 2012, I spent around 2 years leading the scientific work for the relevance of ads and sponsored updates, especially, response prediction. These efforts have significantly improved performance metrics for the two products, e.g. CTR and revenue. I am currently serving as a technical lead of the LinkedIn relevance team (around 180 people) that tackles all the relevance and data related problems at LinkedIn, e.g. homepage feed personalization, site speed data science, email optimization, jobs recommendation, anomaly detection and so forth.

    Selected Awards:
    Yahoo! Super Star Team Award (highest team achievement award in the company) (Abuse team), 2011
    Yahoo! Super Star Team Award (Taiwan Auction team), 2011
    Research Assistantship, Duke University, 2005- 2008
    Teaching Assistantship and Fellowship, Duke University, 2004
    USTC Outstanding Student Scholarship (2000-2003)