Name: The SRE’s Guide to LLMOps: Day-0 Deploying and Managing AI Workloads on Kubernetes
Start: 2025-03-11T14:00:00Z
End: 2025-03-11T14:00:31.000Z
Location: BrightTALK
Rating: 3

Mirantis helps organizations ship code faster utilizing public and private clouds, providing a public cloud experience on any infrastructure from the data center to the edge. With a long-standing record of delivering solutions based on open source with enterprise-grade support, Mirantis offers an as-a-service experience of Kubernetes and OpenStack to provide containerization and virtualization services. Mirantis serves many of the world’s leading enterprises, including Adobe, DocuSign, Liberty Mutual, Nationwide Insurance, Reliance Jio, S&P Global, Seagate, Société Générale, and Volkswagen. Learn more at www.mirantis.com.

Mirantis is a global company headquartered in Campbell, California.

As AI adoption accelerates across governments, regulated industries, and regional cloud providers, the limits of hyperscaler architectures are becoming increasingly visible. Sovereign AI demands strict data residency, tenant isolation, governance, and regional control — requirements traditional cloud and Kubernetes models were never built to meet.

In this session, we break down the core architecture of a true Sovereign AI Cloud. You’ll learn why hyperscaler patterns fall short, what a secure and compliant multi-tenant environment really requires, and how an open, vendor-neutral AI Factory model enables sovereign operations across any infrastructure.

We’ll also walk through the Nebul case study — a real sovereign cloud built to meet European privacy, security, and operational standards. You’ll see how Nebul solved challenges like GPU sprawl, isolation, lifecycle complexity, and multi-cluster scale using k0rdent’s open architecture.

What you’ll learn:
- Why hyperscaler and traditional Kubernetes models fail to meet sovereign AI requirements
- The essential components of a compliant, secure, multi-tenant Sovereign AI Cloud
- How an open AI Factory model enables isolation, governance, and regional autonomy
- How Nebul built a real Sovereign AI Cloud and what you can apply to your own architecture

If you’re planning or evaluating a Sovereign AI initiative in 2026, this session will give you the architectural clarity and blueprint you need to move forward confidently.

How to Build a Sovereign AI Cloud Hyperscalers Can’t Deliver — Proven in the Real World

AI adoption is accelerating across every industry, and 2026 will mark a major shift in how organizations design and operate their infrastructure. Traditional cloud patterns are reaching their limits as enterprises face GPU-intensive workloads, data locality challenges, multi-cluster operations, and stricter security and governance requirements.

This session explores the key AI infrastructure trends defining 2026 and introduces the AI Factory — a standardized, open architectural model that brings consistency to data pipelines, governance, security, and deployment. We’ll walk through the AI Factory Reference Architecture, a vendor-neutral blueprint that supports any GPU, any cloud, and any environment. Finally, we’ll show how k0rdent AI implements this architecture to deliver flexible, efficient, and lock-in-free AI operations at scale.

Key Takeaways:
- The top AI infrastructure trends shaping 2026 and why current models are hitting limits
- What an AI Factory architecture is and why enterprises are adopting it
- How an open, vendor-neutral AI Factory avoids lock-in and scales across any environment
- How k0rdent AI brings this architecture to life for modern AI operations

Reserve your seat to see how an open, Kubernetes-based AI Factory can accelerate your organization’s AI strategy in 2026 and beyond.

Scaling AI in 2026: The Rise of the Open AI Factory on Kubernetes

Building and scaling cloud products shouldn’t require writing complex code or managing endless dependencies. For Cloud Service Providers, meeting growing customer demand while maintaining speed and quality can be a major challenge.

In this 20-minute session, Mirantis’ Anjelica Ambrosio introduces the k0rdent AI Cloud Services Portal  — a technology that empowers Cloud Service Providers to launch GPU-enabled services faster than ever. You’ll see how k0rdent AI lets operators and customers alike build and customize products from pre-rendered components from an extensive catalog, drastically simplifying development and delivery.

What you’ll learn:
- How k0rdent AI’s Cloud Services Portal streamlines product creation and white-labelling
- A live demo of setting up and deploying a GPU-enabled cluster product — from operator to customer view
- How to bring new cloud services to market with ease and speed

If you’re a Cloud Service Provider, Platform Engineer, or Product Owner looking to accelerate time-to-market for GPU-powered offerings, this short session is for you.

GPU Cloud-in-a-Box: Build Products at Ease

Deploying GPU inference workloads can be a complex and fragile process. Manual service installation, dependency management, and configuration errors often slow down teams and delay results. These challenges create friction for operators who want to move fast and scale AI applications efficiently.

This 20-minute webinar, led by Mirantis’ Anjelica Ambrosio, shows how k0rdent AI streamlines GPU inference deployment using templates and automation. You will see how pre-built templates, automated dependency provisioning, and seamless run:ai integration simplify every stage from setup to production, transforming GPU management into a fast, reliable, and repeatable process.

What you’ll learn:
- Why GPU inference workloads are difficult to deploy and manage
- How k0rdent AI uses templates to automate installation and configuration
- How run:ai integration and dependency automation accelerate deployment
- How to transform manual GPU workflows into scalable, production-ready infrastructure

If you are a Platform Engineer, DevOps professional, or IT leader responsible for enabling AI infrastructure, this session gives you a clear path to deploy and scale GPU inference with confidence.

Turning Bare Metal into AI Powerhouses: Deploy, Scale, and Run Smarter with k0rdent AI

Running AI workloads on bare metal servers unlocks the full power of CPUs and GPUs by eliminating virtualization overhead. But provisioning and managing bare metal infrastructure for AI is complex, time-consuming, and error-prone. Teams struggle with setup, resource utilization, and the lack of proper orchestration and monitoring, which slows down innovation and increases costs.

This 20-minute webinar, led by Mirantis’ Anjelica Ambrosio, shows how k0rdent AI simplifies bare metal lifecycle management for AI workloads. You will see how discovery, provisioning, and monitoring are automated, and how servers seamlessly transition from available to in-use, so your team can focus on accelerating AI applications instead of fighting infrastructure.

What you’ll learn:
- Why bare metal is the optimal foundation for AI workloads
- The biggest challenges of provisioning and managing bare metal for AI
- How k0rdent AI automates server lifecycle management from discovery to in-use
- How to streamline infrastructure creation and maximize performance for AI projects

If you are a Platform Engineer, Architect, or IT leader tasked with enabling AI at scale, this session gives you the roadmap to simplify infrastructure and unlock faster results.

Are You Ready for AI? Transform Your Bare Metal Servers with k0rdent AI

SUSE Rancher RKE1 has officially reached end of life. There’s no in-place upgrade path. No clear continuity. Just a complex replatforming mandate and rising security risk. For platform teams running SAP, managing edge environments, or operating in airgapped conditions, RKE3 often introduces more problems than it solves, poor disconnected support, insecure cert handling, and chaotic upgrades that strain teams and systems.

This live roundtable brings together Kubernetes, edge, and airgap experts to unpack what RKE1 EOL really means for your infrastructure. We’ll break down the migration risks, highlight critical vulnerabilities, and show how Mirantis Kubernetes Engine 4k (MKE4k) offers a secure, composable, and Zero-Touch alternative for regulated and edge-heavy environments.

What you'll learn:
• Why RKE1’s EOL presents major risk to regulated industries and edge deployments
• The hidden operational and security costs of moving to RKE3
• How MKE4k simplifies replatforming with airgap-ready automation and Zero Trust architecture

What platform teams need to future-proof Kubernetes strategy across edge, cloud, and airgapped infrastructure
If you’re a Platform Engineer, Architect, or IT leader evaluating life after Rancher, this session is your roadmap forward.

Rancher at the End of the Road: What’s Next for Kubernetes?

AI initiatives often start with promise, but scaling them into production is where most organizations struggle. The result is fragmented teams, underutilized GPU clusters, rising costs, and ad hoc AI projects that cannot be repeated or sustained.

In this webinar, we will explore how forward-thinking enterprises are transforming GPU sprawl into highly efficient AI Factories by building repeatable, scalable AI and ML pipelines. You will learn how to:
• Streamline GPU provisioning and orchestration across hybrid and multi-cloud environments
• Establish standardized, reusable workflows for training, inference, and deployment
• Optimize infrastructure utilization to eliminate bottlenecks and control costs
• Enable platform teams to deliver self-service AI environments for data scientists and developers

This session will provide a blueprint for taking your AI strategy from disorganized experimentation to an industrialized factory model that accelerates innovation and delivers measurable ROI at enterprise scale.

Who should attend: Platform engineers, AI and ML architects, DevOps leaders, and technology decision-makers responsible for enterprise AI adoption

From GPU Chaos to AI Factory: How Enterprises are Building Repeatable AI Pipelines at Scale

Sovereign cloud providers and regional telcos face mounting pressure to deliver scalable, profitable AI infrastructure without the complexity or cost of hyperscaler stacks. Traditional IaaS models were not designed for the elasticity, scale, or GPU efficiency that modern AI workloads demand. As a result, many providers experience sluggish customer onboarding, idle GPU capacity, and missed monetization opportunities.

In this webinar, you will learn how to turn your GPU infrastructure into a high-margin AI service platform. Discover how providers are using k0rdent AI’s turnkey deployment templates to launch GPU-backed services in days, reduce time to market, optimize GPU usage, and eliminate infrastructure waste.

To bring the ROI story to life, we will showcase live OpenCost-Grafana dashboards and forecasting metrics that highlight cost efficiency, revenue impact, and long-term profitability. You will walk away with practical insights to help you build a scalable and sovereign AI cloud strategy.

What you’ll learn:
• Why traditional cloud infrastructure falls short for AI workloads and what to do about it
• How to accelerate customer onboarding with k0rdent AI’s GPU service templates
• How to maximize GPU utilization and forecast ROI with OpenCost-Grafana dashboards

This session is ideal for Heads of Cloud Services, Product VPs, and CTOs, and also valuable for anyone involved in building AI infrastructure or launching GPU-backed services in sovereign, regulated, or regional markets.

How Sovereign Cloud Providers Can Monetize GPU Infrastructure Without Hyperscaler Complexity

Scaling AI across edge, core, and cloud environments is now a top priority but also a top challenge. How do you deliver high-performance inference, maintain control over your data, and keep costs predictable across a hybrid or multi-cloud setup?

In this session, you'll learn how to design and operationalize GPU-powered infrastructure that meets enterprise-grade AI demands. Whether you're leading platform architecture or responsible for AI/ML infrastructure strategy, you'll walk away with:
• Proven patterns for building hybrid GPU clouds that support both training and inference
• A framework for balancing performance, cost, and data sovereignty
• Strategies to enable inference-as-a-service across teams and regions
• Insights into how Kubernetes-native orchestration and NVIDIA technologies can accelerate deployment at scale

Led by Senior DevOps Engineer Prashant Ramhit, this session offers actionable guidance based on real-world deployments. If you're building or scaling intelligent applications, this is a can't-miss discussion on how to do it right.

AI at the Edge, Core, and Cloud: Building GPU Clouds for the Next Era of Intelligent Applications

As AI adoption accelerates, sovereign cloud providers and regulated enterprises face increasing pressure to offer scalable AI services while maintaining full control over data, infrastructure, and compliance. However, building secure, multi-tenant GPU infrastructure across hybrid environments remains a major operational challenge.

In this session, you’ll learn how k0rdent AI helps organizations streamline GPU provisioning, and deploy AI workloads at scale. The platform is built on a Kubernetes-native architecture designed for high-compute, security-sensitive environments.

We’ll also explore a real-world case study featuring Nebul, a sovereign cloud provider based in the Netherlands, and show how they operationalized their GPU clusters to serve sensitive AI workloads securely and efficiently.

What you’ll learn:
- How to automate GPU-ready Kubernetes clusters with compliance in mind
- Best practices for deploying and scaling multi-tenant AI inference services using KServe, Knative, and Istio
- Lessons from Nebul’s journey as a sovereign CSP scaling secure AI infrastructure

Whether you’re building sovereign cloud services or modernizing infrastructure for sensitive AI workloads, this session will give you the blueprint to do it securely, efficiently, and at scale.

Sovereign AI Infrastructure at Scale: Real-World Blueprint for Cloud Providers

Managing Kubernetes clusters across cloud, on-prem, and edge environments introduces challenges in scalability, networking, and security.

In this joint webinar, Mirantis and Calico experts  will demonstrate how the open source tools k0rdent and Calico work together to simplify multi-cluster operations and enhance network visibility and protection.

You’ll learn how Mirantis k0rdent makes it easier to provision and manage Kubernetes clusters across AWS and Azure, with Calico automatically deployed as the CNI. k0rdent simplifies multi-cluster operations by centralizing lifecycle management and policy enforcement, helping teams maintain consistency across diverse environments with less manual effort.

You’ll also explore how Calico provides deep visibility into Kubernetes network traffic, strengthens network security, and improves overall reliability. The session will cover how Calico supports scalable, secure networking—and how organizations with more advanced needs can extend to Calico Enterprise for unified network security management across hybrid and multi-cloud deployments.

Whether you're running a few clusters or operating at global scale, this session will share practical strategies to simplify operations and strengthen your Kubernetes networking posture.

Simplify and Secure Multi-Cluster Kubernetes with k0rdent and Calico

As Kubernetes environments grow across clouds and data centers, ensuring consistent security, compliance, and operational policies becomes critical — and increasingly complex.

Join Mirantis and Nirmata to see how open source k0rdent and Kyverno make policy-driven Kubernetes management simple, scalable, and automated across multi-cluster environments.

In this session, you’ll discover how:
- Discover how k0rdent simplifies creating and managing secure multi-cluster Kubernetes environments across clouds and data centers.
- Learn how Kyverno automates policy-driven governance, compliance enforcement, and security across distributed Kubernetes clusters.
- See how k0rdent and Kyverno together simplify operations, enable secure developer self-service, and accelerate cloud-native initiatives.

Whether you're building internal developer platforms (IDPs), supporting AI-driven workloads, or securing large-scale Kubernetes deployments, this webinar will show you how to deliver secure, compliant, and scalable Kubernetes environments — with less complexity and more control.

Policy-Driven Multi-Cluster Kubernetes for AI Workloads Made Simple with k0rdent and Kyverno

Deploying AI on Kubernetes can be complex, involving GPU provisioning, model serving, and cost-efficient scaling. This session explores how to streamline AI deployments on Kubernetes using k0rdent. 

Attendees will learn how to automate GPU ready cluster provisioning, deploy scalable AI models with Kserve, and optimize inference workloads with Knative and Istio. With GPUs being constrained and costly, this session emphasizes how auto-scaling can maximize resource utilization while minimizing operational costs.

The talk will also cover GPU monitoring strategies using Prometheus and Grafana to ensure optimal performance. Join us to see how k0rdent empowers AI teams to deploy, scale, and optimize models effortlessly—ensuring maximum performance at the lowest cost.

AI on Kubernetes: Scaling Smarter, Running Faster with k0rdent

As financial institutions modernize their IT infrastructure, traditional virtualization solutions often impose high costs, operational complexity, and vendor lock-in. Cloud-native virtualization offers a scalable, cost-efficient alternative that enhances agility, security, and compliance—critical factors for financial services innovation.

In this webinar, we’ll explore a real-world use case of cloud-native virtualization in the financial sector. Our speaker, Michael Tillman, Solution Architect Manager, will discuss how a leading financial institution successfully transitioned from legacy virtualization to a Kubernetes-powered, cloud-native approach. 

You'll gain insights into:
- The challenges of traditional virtualization in financial services
- How cloud-native virtualization optimizes performance and reduces costs
- Security, compliance, and operational benefits for regulated environments
- Lessons learned and best practices for IT teams in banking and financial services

Join us to discover how financial institutions can achieve greater flexibility and efficiency with cloud-native virtualization. Don't miss this opportunity to learn from real-world experiences and get your questions answered by cloud native experts!

From Legacy to Cloud-Native: Virtualization Strategies for Financial Institutions

Managing Kubernetes at scale can be complex, with operational overhead and vendor lock-in limiting flexibility and innovation. Open-source solutions like k0rdent empower enterprises with a vendor-neutral, scalable, and efficient approach to Kubernetes management.

Join us for an insightful webinar where we explore how k0rdent, an open-source Distributed Container Management Environment (DCME), provides a vendor-neutral, scalable, and efficient approach to multi-cluster Kubernetes management—giving platform engineers and IT leaders the control, automation, and flexibility they need.

What You’ll Learn:
- Break Free from Vendor Lock-In: Discover strategies to build a flexible, open Kubernetes ecosystem.
- Simplify Multi-Cluster Management: Step-by-step guidance on managing Kubernetes at scale with reduced complexity.
- Optimize Workloads for Efficiency: Real-world use cases of organizations leveraging k0rdent to enhance performance and lower costs.
- Seamless Integration with Existing Operations: Actionable insights on incorporating k0rdent into your infrastructure for improved automation and scalability.

Whether you’re a platform engineer, DevOps leader, or IT decision-maker, this session will provide practical knowledge to elevate your Kubernetes strategy.

Don’t miss this opportunity to take control of your Kubernetes environment with open-source innovation. Register now and take the next step toward a more agile, scalable future with k0rdent.

The Case for Open Source: How k0rdent Empowers Enterprise Agility

Modern governments face growing demands to modernize IT systems, improve scalability, and deliver secure, efficient services. Cloud-native virtualization offers a streamlined path to transformation, enabling governments to overcome challenges and build future-ready infrastructure.

Join us for a webinar as we explore a real-world use case that demonstrates the benefits of adopting cloud-native technologies in a government setting.

You’ll learn:
- How cloud-native virtualization reduced operational complexity.
- Strategies to achieve scalability and improve service efficiency.
- Real-world insights into overcoming legacy system challenges.
- Best practices for implementing secure, future-ready infrastructure.

Gain actionable insights into overcoming modernization challenges and building future-ready infrastructure solutions.

Cloud Native Virtualization in Government: A Real-World Use Case

Kubernetes platform engineering is complex, especially with large, shared clusters supporting multiple workloads and teams.

Defeating complexity requires multi-cluster thinking: Compose clusters declaratively with services and policies to support one or a small range of workload types (versus many); administer and lifecycle manage declaratively to continually assure cluster state; and repeat for all the workload types your organization needs to support.

Simplifying platforms and customizing them to workload requirements pays off bigtime in agility, smoother operations, higher workload availability, security, and compliance. Plus: You don't need to implement a global multi-cluster architecture to do this at smaller scales.

In this webinar, we’ll explore:
- How to think of Kubernetes platforms as containers for workloads, services (LMA, FinOps), and policies.
- Building up use-case-optimized platforms from 'bare clusters.'
- Approaches to automate platform operations and LCM.
- Strategies to keep platforms healthy and integrated over time.

And we'll share how we're doing all of this with open-source-based tools. Join us!

Defeat Complexity with Use-Case-Optimized Kubernetes Platforms

For SRE or DevOps professionals navigating uncharted waters of LLMs, 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. 

Join our webinar led by Harsh Mishra, Site Reliability Engineer at One2N Consulting, to explore the unique challenges and solutions of LLMOps from an SRE perspective.

What we’ll cover:
- Day 0 prerequisites for understanding and hosting LLMs on Kubernetes.
- Breaking down core components: AI (RAGs, Vector Databases), Backend, and Compute.
- Key differences between LLMOps and traditional MLOps.
- Case study: Building an internal tool with optimized architecture using Nvidia GPUs and Ray Distributed.
- Lessons learned in our journey in helping organisations build the necessary GenAI muscle for SREs.

Register now!

The SRE’s Guide to LLMOps: Day-0 Deploying and Managing AI Workloads on Kubernetes

Kubernetes

LLMOps

DevOps

Welcome to the virtualization community on BrightTALK! Whether it affects servers, storage, networks, desktops or other parts of the data center, virtualization provides real benefits by reducing the resources needed for your
infrastructure and creating software-defined data center components. However, it can also complicate your infrastructure. Join this active community to learn best
practices for avoiding virtual machine sprawl and other common virtualization pitfalls as well as how you can make the most of your virtualization environment.

Virtualization

Cloud computing has exploded over the past few years, delivering a previously unimagined level of workplace mobility and flexibility. The cloud computing community on BrightTALK is made up of thousands of engaged professionals learning from the latest cloud computing research and resources. Join the community to expand your cloud computing knowledge and have your questions answered in live sessions with industry experts and vendor representatives.

Cloud Computing

The IT security community on BrightTALK is composed of more than 200,000 IT security professionals trading relevant information on software assurance, network security and mobile security. Join the conversation by watching on-demand and live information security webinars and asking questions of experts and industry leaders.

IT Security

The data center management community focuses on the holistic management and optimization of the data center. From technologies such as virtualization and cloud computing to data center design, colocation, energy efficiency and monitoring, the BrightTALK data center management community provides the most up-to-date and engaging content from industry experts to better your infrastructure and operations. Engage with a community of your peers and industry experts by asking questions, rating presentations and participating in polls during webinars, all while you gain insight that will help you transform your infrastructure into a next generation data center.

Data Center Management

The application development community features top thought leadership focusing on optimal practices in software development, SDLC methodology, mobile app development and application development platforms and tools. Join top software engineers and coders as they cover emerging trends in everything from enterprise app development to developing for mobile platforms such as Android and iOS.

Application Development

Network infrastructure professionals understand that a reliable and secure infrastructure is crucial to enabling business execution. Join the network infrastructure community to interact with thousands of IT professionals. Browse hundreds of on-demand and live webinars and videos to learn about the latest trends in network computing, SDN, WAN optimization and more.

Network Infrastructure

As an IT professional, many of the problems you face are multifaceted, complex and don’t lend themselves to simple solutions. The information technology community features useful and free information technology resources. Join to browse thousands of videos and webinars on ITIL best practices, IT security strategy and more presented by leading CTOs, CIOs and other technology experts.

The SRE’s Guide to LLMOps: Day-0 Deploying and Managing AI Workloads on Kubernetes

Presented by

About this talk

Mirantis