CPU vs GPU // Ronen Dar & Gijsbert Janssen van Doorn // MLOps Coffee Sessions #99

MLOps Coffee Sessions #99 with Ronen Dar and Gijsbert Janssen van Doorn, Getting the Most Out of your AI Infrastructure co-hosted by Vishnu Rachakonda.   // Abstract Run:AI is building a cloud-based platform for building with AI. In this talk, we hear all about why this need exists, how this works, and what value it creates. // Bio Ronen Dar Run:AI Co-founder and CTO Ronen was previously a research scientist at Bell Labs and has worked at Apple and Intel in multiple R&D roles. As CTO, Ronen manages research and product roadmap for Run:AI, a startup he co-founded in 2018. Ronen is the co-author of many patents in the fields of storage, coding, and compression. Ronen received his B.S., M.S., and Ph.D. degrees from Tel Aviv University. Gijsbert Janssen van Doorn Gijsbert is Director of Technical Product Marketing at Run:AI. He is a passionate advocate for technology that will shape the future of how organizations run AI. Gijsbert comes from a technical engineering background, with six years in multiple roles at Zerto, a Cloud Data Management and Protection vendor. // MLOps Jobs board   https://mlops.pallet.xyz/jobs // Related Links The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers by Ben Horowitz ebook:  https://www.scribd.com/book/211302755/The-Hard-Thing-About-Hard-Things-Building-a-Business-When-There-Are-No-Easy-Answers?utm_medium=cpc&utm_source=google_search&utm_campaign=3Q_Google_DSA_NB_RoW&utm_term=&utm_device=c&gclid=Cj0KCQjw1ZeUBhDyARIsAOzAqQLnUzXlgFT1PjU_M6jGqRZmwLbcK-mbfKQI4XrZJBRwgUs4x5j2hQ4aAmt1EALw_wcB --------------- ✌️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 Vishnu on LinkedIn: https://www.linkedin.com/in/vrachakonda/ Connect with Ronen on LinkedIn: https://www.linkedin.com/in/ronen-dar/ Connect with Gijsbert on LinkedIn: https://www.linkedin.com/in/gijsbertjvd/ Timestamps: [00:00] Introduction to Ronen Dar & Gijsbert Janssen van Doorn [01:25] Takeaways [04:24] Thank you Run:AI for sponsoring this episode! [05:13] Run:AI products and components [09:27] Companies coming to Run:AI and problems they solve [13:30] Why is this problem hard? [18:56] Run:AI's Vision [22:12] Run-on-the-mill workload [25:36] Engineering challenges and requirements building Run:AI   [32:47] Process of solving problems on the same page [35:45] Power to give data scientists [37:38] Avoiding horror stories that might cost a lot of money [44:23] Running multiple models on a single GPU [47:17] Never scale down to zero [48:28] So many ML Start-ups in Israel [53:00] Vision for the future at GPUs and how will Kubernetes advance [55:55] Future of AI accelerators [57:03] Lightning round [1:02:26] Wrap up

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.