Jibin Zhan is the Vice President of the Platform and Product Engineering at Conviva, a startup that helps top broadcasters, operators, and content owners experience excellence and create more profitable streaming businesses with its Intelligent Control Platform. He is responsible for Conviva’s big data computation and infrastructure, online video monitoring and optimization, web applications, data analysis and data science teams. Before Conviva, he worked for several years at Carnegie Mellon University, doing research in the area of Internet Infrastructure and Internet-based applications, such as 100×100, End System Multicast projects. He has over 20 years of experience in senior technical management and senior engineer positions at Conviva, Turin Networks, Cisco, UUNET, and Nortel, etc.
Conviva is helping hundreds of premium content owners, broadcasters and service providers around the world to optimize their user's video experience by monitoring, managing and optimizing billions of streams every month. In this talk, we are going to discuss how we use Spark as one unified framework and a single code base to provide real-time, near real-time and offline analysis for the online video business of our customers. In particular, we provide our customers services such as real-time analytics, real-time alerts, diagnosis, automatic video optimization, as well as more sophisticated offline business analysis, such as video experience benchmarking, root cause analysis, and subscriber retention analysis.. We will explain how Databricks has dramatically improved the efficiency of our data analysis and data science team, with an environment that is powerful, flexible, and cost-effective, enabling rapid product prototyping and iterations at Conviva.