A SRE’s Guide to LLMOps: Deploying and Managing AI/ML Workloads using Kubernetes скачать в хорошем качестве

A SRE’s Guide to LLMOps: Deploying and Managing AI/ML Workloads using Kubernetes

Learn more about Mirantis at https://www.mirantis.com/ Learn more about One2N at https://one2n.io/ For Site Reliability Engineers (SREs) or DevOps professionals navigating the uncharted waters of Large Language Models (LLMs), Retrieval-Augmented Generation (RAGs), and GenAI, the lack of standardization, resources, and knowledge can quickly turn into an operational nightmare. DevOps and SRE practices are no longer just about infrastructure and delivery speed - they now also delve into the complexities of GenAI, particularly LLMs. While it may be tempting for an uninitiated SRE to approach this as a traditional MLOps problem, it quickly becomes clear that LLMOps is an entirely different challenge, starting from Day 0. This webinar led by Harsh Mishra, SRE at One2N Consulting, explores the unique challenges and solutions of LLMOps from an SRE perspective. The best practices he outlines will provide viewers with a great jumping off point for LLMOps, as well as future-proofing their cloud strategy to include the rapidly evolving field of GenAI and LLMs. What this presentation will cover: ▫️ Day 0 prerequisites for understanding & hosting LLMs on Kubernetes ▫️ Core components - AI (RAGs/Vector Databases), Backend & Compute ▫️ LLMOps vs traditional MLOps - key differences ▫️ Case study - build an internal tool using Nvidia GPUs & Ray Distributed ▫️ Lessons learned by helping Orgs build the necessary GenAI muscle for SREs This webinar is a necessary first step in understanding GenAI as it relates to the cloud and Kubernetes, as well as helping to differentiate between established MLOps and the operations specific to Large Language Models and Retrieval-Augmented Generation. Whether you’re an SRE beginning your journey with LLMs and RAGs, a DevOps professional looking for ways to streamline your work with GenAI, or you are just interested in the latest developments in GenAI and LLMs, this webinar will leave you with a more comprehensive understanding of GenAI’s current positioning in the cloud native and Kubernetes space. 📖 For more on the intersection of AI/ML and Kubernetes, check out our recent blog: https://www.mirantis.com/blog/like-sw... 🔬 If you are interested in more Tech Talks from the Mirantis Labs team, on demand recordings as well as upcoming presentations can be found at https://www.mirantis.com/labs/ ⚙️ Access the Mirantis Labs GutHub Repo for helpful resources: https://github.com/mirantis-labs 📚 If you would like more informational resources about Mirantis, our solutions & related areas of expertise, please check out our Resource Library: https://www.mirantis.com/resources/ #kubernetes #k8s #devops #machinelearning #ai #aiops #llms #llmops Get started with our products: https://www.mirantis.com/get-started/ World class training from Mirantis: https://training.mirantis.com/ Connect with us on LinkedIn:   / mirantis   Follow us on Twitter:   / mirantisit   Friend us on Facebook:   / mirantisus   CHAPTERS 0:00 - Intro 0:24 - Why SREs need to worry about LLMs 1:11 - One2N's journey towards true LLMOps 15:30 - One2N's successes & challenges with K8s LLMOps 18:35 - MLOps vs LLMOps: Infra, Deployment & Architecture 19:31 - VMs to Kubernetes for LLMOps 23:07 - Vector Storage: index vs database 25:12 - Deep dive into One2N's Lab setup (real-world use case) 28:58 - Key takeaways to get started with LLMOps 31:23 - Outro