Paavany is a Senior PM in the Office 365 Customer Fabric team at Microsoft, whose mission is to attract, retain, and engage users with the help of their big data products. Prior to this, Paavany worked on PowerMap, a 3D data visualization and storytelling tool, taking it from an incubation project into production. Before that she worked on Microsoft Excel on features like PivotTables and Conditional Formatting.
Keeping millions of paying users happy - how the Office 365 Customer Fabric helps us attract, retain and engage users. Office 365 is one of the world's biggest subscription services-expectations are high as customers trust us with their most critical documents, communications and workflows. Keeping everyone happy and creating opportunities for us to delight our users is no easy task. We've leveraged Cassandra and Spark among other systems to build our "Customer Fabric", where inferences are generated to drive outreach, satisfaction, upselling, acquisition as well as predictive user interfaces. We believe deep user understanding is strategic and key to sustained growth and engagement. To do this, the Office 365 Customer Fabric team deployed Apache Spark, Apache Kafka, and Cassandra in their production environment. With so much data velocity with high-frequency data and 10TB stored a day, they needed a platform that could handle their diverse big data requirements. In this session, we will discuss how we used Spark to solve their need for both real-time and batch mode analytics. This includes utilizing user and customer level aggregations, MLlb techniques for ticket classification, and real-time ingestion of data.