Technical Talks

View All

The Data Infra Behind Zillow's 3x Growth in Experiment Volume

Aaron Wroblewski Aaron Wroblewski | Senior Manager, Machine Learning Engineering & Science | Zillow

At Zillow, we've implemented a cutting-edge data infrastructure that has enabled us to increase our experiment volume by 3x. This talk will give you a deep dive into the key components of this infrastructure, including our flexible experiment analysis engine, single-source of truth API, and the Kafka and Kubernetes infra powering eventing and enabling rapid development. We've achieved significant cost and time savings, enabling us to ship faster and compute more results to support more decisions more efficiently. Join us to learn how we've streamlined our experimentation process thanks to data infra and see a real-world example of the impact of this infrastructure. This talk is for data engineers, data scientists, and technical decision-makers who want to increase speed of developing data applications.

Aaron Wroblewski
Aaron Wroblewski
Senior Manager, Machine Learning Engineering & Science | Zillow

Aaron is a seasoned data, engineering, and science leader with a passion for building cutting-edge experimentation and machine learning solutions. At Zillow, he leads a team focused on enabling faster, more trustworthy decisions and growing value for customers, partners, and the business. Our team, ZEXP, is building Zillow’s first end-to-end self-service experimentation platform, and conducting research to increase quality and speed of decision making, through variance reduction, meta-analysis, and behavior modification. Previously at Zillow, Aaron and his team completed pioneering work in user intent modeling, contextual bandit UX optimization, and applying machine learning to marketing spend optimization and personalization. With a rich background in experimentation & machine learning at Amazon Search and developer tools at Microsoft, Aaron brings a unique perspective to data-driven product development.