Presented by:Sho Soboyejo
Machine Learning is on another hype cycle with generative interfaces such as ChatGPT and MidJourney pushing the edge on NLP and Image generation challenges. However, there’s more to deploying ML systems than the modeling technique or the UX. What ML design systems considerations do you need to tackle classical problems in the domain of search, recommendation systems or ads prediction? How do you confidently deploy your models to production?
As an ML Engineer you need to be able to build distributed systems that address scalability, observability and other NFRs. You need solid MLOps techniques to deliver true value to your users.
This talk dives into five critical ML system design patterns, illuminating methodical approaches to traditional ML business problems. You'll acquire a solid understanding of essential MLOps practices necessary for scaling your models and responding to production challenges.
Join us as we unravel the intricacies of ML system design and MLOps, empowering you to implement scalable and adaptive ML solutions in production. This isn't just a technical walkthrough - it's a roadmap to enhancing business problem-solving and user value delivery using Machine Learning.
Level: IntermediateTags:AI & ML, Data, Patterns & Practices