Facebook has nearly a billion users generating massive amounts of structured content on a regular basis. Given the scale of content creation, we needed an approach to build, maintain, and utilize the object graph.
This session explores the framework Justin’s team developed from bootstrap to the long tail, with parallel concrete examples from Facebook’s location and places effort. Justin will describe the most effective dataset management techniques, including a combination of product, crowdsourcing, machine learning and search ranking. He will also cover some of the products his team built using a unique structured dataset and how you can do the same by leveraging the custom open graph.
With one engineer per million active users, Facebook focuses on developing efficient, impactful, and high leverage solutions to its problems. These techniques should therefore apply to even the most resource-constrained companies.
Justin is a member of the Entities/Data Science team at Facebook where he helps curate and build from their rich structured object and social graphs, with a focus on location. Before joining Facebook, Justin ran the Data team at foursquare. In addition to building their core data-driven products Explore and Radar, he built a team from the ground up that consisted of Engineers and Data Scientists to solve large scale data problems as foursquare’s dataset grew from half a million check-ins to over 1.5 billion. Before that, Justin worked at a hedge fund as a quantitative analyst, building custom portfolios for their asset management division and doing modeling and analysis for their risk team, specializing in high-frequency, derivatives, and commodities trading. Prior to that, he worked for Bear Stearns as a Vice President in their fixed income analyst group, building applications and models to help value agency pass-thru securities and building loan-level pricing applications and models. Justin holds a BS in Computer Science with a minor in Mathematics from the University of Rochester and has studied graduate-level Math and Computer Science at Columbia University. He is constantly chasing the biggest and most interesting datasets and trying to make amazing things happen with them.
For information on exhibition and sponsorship opportunities, contact Susan Stewart at firstname.lastname@example.org.
For information on trade opportunities contact Kathy Yu at mediapartners
For media-related inquiries, contact Maureen Jennings at email@example.com
View a complete list of Strata contacts.