Scheduled for: Interviews/ Category:
How can ML technology help manage insane volumes of data and keep cloud costs in check?
Summary of the video
Barzan Mozafari, the founder and CEO of Kibo, discusses the role of machine learning in helping data teams manage large volumes of data and control cloud costs. The conversation covers various aspects of Kibo’s approach and its benefits for businesses. Barzan highlights the challenges faced by data teams in handling massive data volumes and the increasing complexity of data pipelines. He explains that while cloud data warehousing solutions have lowered the adoption barrier, they have led to higher operating costs.
Barzan discusses how Kibo’s platform uses machine learning to automatically optimize data pipelines and workloads. He emphasizes that Kibo’s goal is not to replace human experts but to alleviate them from manual tasks so they can focus on asking the right questions and making business decisions. The platform continuously monitors and learns from the customer’s workload, offering suggestions, trends, and root causes for various phenomena. Kibo’s approach is to provide a trial period during which customers maintain control and visibility over the optimization process.
Barzan shares insights about founding a company, highlighting the importance of putting customers and teammates first. He stresses the significance of listening to customers’ pain points and feedback, which has been crucial in the success of Kibo. He mentions that Kibo’s platform does learn from customer interactions and data to enhance its capabilities but also emphasizes the value of direct communication with customers.
Barzan has spent the last 15 years of his career on statistical and machine learning techniques for building smarter, faster, more scalable data-intensive systems. He is an entrepreneur, researcher, and professor.
Alongside being Founder and CEO of Keebo, providing fully-automated data warehouse and analytics optimization solutions, he is an Associate Professor of Computer Science and Engineering at the University of Michigan, where he leads a research group designing the next generation of scalable databases using advanced statistical models.
His research career has led to several major open-source projects, and he has won the National Science Foundation CAREER award, as well as several best paper awards in ACM SIGMOD and EuroSys.