Feathr: LinkedIn's High-performance Feature Store // David Stein // Coffee Sessions #120

MLOps Coffee Sessions #120 with David Stein, Senior Staff Software Engineer at LinkedIn, Feathr: LinkedIn's Enterprise-Grade, High-Performance Feature Store co-hosted by Skylar Payne. // Abstract When David started building Feathr, Feature Stores did not exist. That was not a term floating around at all. This was definitely one of the OG Feature Stores for sure!   We hear how the LinkedIn team got to this point, having an open source release, and how they used LinkedIn as an incubator to build a great product. // Bio David Stein is a tech lead at LinkedIn working on machine learning feature infrastructure. He is the original architect of Feathr and continues to contribute to its development as well as to other parts of LinkedIn's machine learning platform. // MLOps Jobs board   https://mlops.pallet.xyz/jobs MLOps Swag/Merch https://mlops-community.myshopify.com/ // Related Links https://engineering.linkedin.com/blog/2022/open-sourcing-feathr---linkedin-s-feature-store-for-productive-m * https://github.com/linkedin/feathr * https://engineering.linkedin.com/blog/2022/open-sourcing-feathr---linkedin-s-feature-store-for-productive-m * https://azure.microsoft.com/en-us/blog/feathr-linkedin-s-feature-store-is-now-available-on-azure/ --------------- ✌️Connect With Us ✌️ ------------- Join our slack community: https://go.mlops.community/slack Follow us on Twitter: @mlopscommunity Sign up for the next meetup: https://go.mlops.community/register Catch all episodes, blogs, newsletters, and more: https://mlops.community/ Connect with Demetrios on LinkedIn: https://www.linkedin.com/in/dpbrinkm/ Connect with Skylar on LinkedIn: https://www.linkedin.com/in/skylar-payne-766a1988/ Connect with David on LinkedIn: https://www.linkedin.com/in/steindavidj/ Timestamps: [00:00] Introduction to David Stein [00:30] Takeaways [06:58] David and Skylar's reunion [08:34] Feathr's background [12:04] Lessons learned in building Feathr [16:35] Scale of Feathr [20:25] Systems interaction [24:15] Standardization [29:20] Importance and better difference of Feathr [34:30] Feature Stores' evolution and more generally MLOps [37:57] Challenges in real-time [40:09] Going real-time, you're not ready! [42:24] Use cases strong for real-time unleveraged now [45:50] Proud about [47:58] Going back in time [50:47] Feathr as a separate company? [52:53] LinkedIn is hiring!

Om Podcasten

Weekly talks and fireside chats about everything that has to do with the new space emerging around DevOps for Machine Learning aka MLOps aka Machine Learning Operations.