Abstract: 

In this session, we will dive deep into Feathr, taking you on a journey into this scalable open-source feature store which has now joined the Linux Foundation AI and Data ecosystem. Feathr has been battle-tested in LinkedIn powering high scale ML applications, supporting 100s of training and inferencing pipelines. This enables feature sharing among teams, leading to significant business metrics gain.

We will dive into some of the key highlights – rich UDF support, dynamic type casting, point-in-time joins, time aware sliding window aggregation, support for derived features, support for advanced ML scenarios and much more!

Feathr has a cloud-friendly scalable architecture and has been made available as an easy deployment on Azure. We will go over the key components – a central registry to store and share feature definitions, offline and online data store connectors, tight integration with Spark to run transformations and integration with various services using the Python SDK.

We will also showcase how Feathr can be used to build an end-to-end solution and go over some of the key customer patterns based on real life customer usage.

Bio: 

Dr. Inchiosa’s passion for AI drives his work as Principal Data Scientist Manager in Azure Data’s Advanced Workload Engineering team, where he leads a team of data scientists focused on AI-led co-innovation engagements with strategic customers and partners. Previously, Mario served as Revolution Analytics’ Chief Scientist and as Analytics Architect in IBM’s Big Data organization, where he worked on advanced analytics in Hadoop, Teradata, and R. Prior to that, Mario was US Chief Scientist in Netezza Labs, bringing advanced analytics and R integration to Netezza’s SQL-based data warehouse appliances. He also served as US Chief Science Officer at NuTech Solutions, a computer science consultancy specializing in simulation, optimization, and data mining, and Senior Scientist at BiosGroup, a complexity science spin-off of the Santa Fe Institute. Mario holds Bachelor’s, Master’s, and PhD degrees in Physics from Harvard University. He has been awarded four patents and has published over 30 research papers, earning Publication of the Year and Open Literature Publication Excellence awards.