Rev 3 Keynote: Cassie Kozyrkov, Chief Decision Scientist at Google

Cassie Kozyrkov addresses the challenges of making data science useful, defining data science as the discipline of making data useful, yet acknowledging the struggle many data scientists face in adding value. She identifies three core problems: the makers, the data, and the definition of "useful." Kozyrkov uses a kitchen analogy to explain the roles within data science, distinguishing between research-focused machine learning engineers who build the "appliances" (algorithms) and applied data scientists who create "recipes" (models) for business problems. She emphasizes the importance of data quality, advocating for a dedicated discipline to ensure data excellence and usability. Kozyrkov further explores the varying excellences of analysts (speed), machine learning engineers (performance), and statisticians (rigor), highlighting the need for skilled decision-makers to steer data science projects effectively.