<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns:yt="http://www.youtube.com/xml/schemas/2015" xmlns:media="http://search.yahoo.com/mrss/" xmlns="http://www.w3.org/2005/Atom">
 <link rel="self" href="http://www.youtube.com/feeds/videos.xml?channel_id=UCdhKM1VoQLhP261oSm79LKA"/>
 <id>yt:channel:dhKM1VoQLhP261oSm79LKA</id>
 <yt:channelId>dhKM1VoQLhP261oSm79LKA</yt:channelId>
 <title>Cloud Native Summit Munich</title>
 <link rel="alternate" href="https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA"/>
 <author>
  <name>Cloud Native Summit Munich</name>
  <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
 </author>
 <published>2024-07-29T06:11:39+00:00</published>
 <entry>
  <id>yt:video:X9OH76DK6H8</id>
  <yt:videoId>X9OH76DK6H8</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>3+1 ways to do MlOps with Kubernetes - Kateryna Hrytsaienko</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=X9OH76DK6H8"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:54:02+00:00</published>
  <updated>2026-10-04T01:28:30+00:00</updated>
  <media:group>
   <media:title>3+1 ways to do MlOps with Kubernetes - Kateryna Hrytsaienko</media:title>
   <media:content url="https://www.youtube.com/v/X9OH76DK6H8?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i1.ytimg.com/vi/X9OH76DK6H8/hqdefault.jpg" width="480" height="360"/>
   <media:description>What happens with models after they are trained? &quot;Whatever they were meant for,&quot; you would say, and you would be completely right. However, what also awaits around the corner are tons of questions like: How to deploy? Where to deploy? How to scale? And, of course, how to limit the budget to a reasonable paycheck?

And here it comes: K8s could save our lives again, and in this session, we will explore 3 and 1 additional K8s-based tools to manage your MLs effectively! All of them we will test against a custom LLM model for performance, scalability and user-friendly deploy process of course!

Speaker: Kateryna Hrytsaienko
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="3" average="5.00" min="1" max="5"/>
    <media:statistics views="167"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:-fF-aoDc2To</id>
  <yt:videoId>-fF-aoDc2To</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>The Future of AI Inference is Small, Distributed, and Budget-Conscious - Mohamed Belgaied Hassine</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=-fF-aoDc2To"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:49+00:00</published>
  <updated>2026-10-15T05:06:18+00:00</updated>
  <media:group>
   <media:title>The Future of AI Inference is Small, Distributed, and Budget-Conscious - Mohamed Belgaied Hassine</media:title>
   <media:content url="https://www.youtube.com/v/-fF-aoDc2To?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i2.ytimg.com/vi/-fF-aoDc2To/hqdefault.jpg" width="480" height="360"/>
   <media:description>By 2030, 90% of AI workloads will be inference—mostly smaller models at smaller organizations. Yet KubeCon talks focus on massive GPU clusters for frontier labs. We're building infrastructure for the wrong market.

This lightning talk argues for a shift: make AI inference accessible to budget-conscious organizations using CNCF projects we already have. I'll sketch a vision combining Cluster API (elastic infrastructure), HAMi (GPU sharing), and Kaito (optimized serving) to create cost-effective, distributed inference for the 90% of organizations who can't afford hyperscaler patterns.

Let's stop optimizing for OpenAI and Anthropic, and start optimizing for everyone else.

Speaker: Mohamed Belgaied Hassine
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="0" average="0.00" min="1" max="5"/>
    <media:statistics views="41"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:VcY_0b0lchk</id>
  <yt:videoId>VcY_0b0lchk</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>In Short: Distributed by DNA: Why Europe's Cloud Market Will and Probably Should Be Decentralized</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=VcY_0b0lchk"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:49+00:00</published>
  <updated>2026-10-15T04:44:53+00:00</updated>
  <media:group>
   <media:title>In Short: Distributed by DNA: Why Europe's Cloud Market Will and Probably Should Be Decentralized</media:title>
   <media:content url="https://www.youtube.com/v/VcY_0b0lchk?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i3.ytimg.com/vi/VcY_0b0lchk/hqdefault.jpg" width="480" height="360"/>
   <media:description>Europe is inherently distributed—in its languages, governments, and laws—meaning our cloud ecosystem shouldn't blindly mimic centralized US hyperscalers. In this rapid-fire session, we will explore why our true potential for digital sovereignty lies in a model matching our DNA: a distributed cloud architecture spanning from federated datacenters down to the edge.

We will quickly highlight how EU tech leadership relies on breakthrough innovations across these distributed clouds, utilizing Kubernetes as the core orchestration layer. Because time is short, we will rapidly synthesize the goals of vital initiatives like IPCEI-CIS, NeoNephos, 8RA, Sovereign Cloud Stack, Eurostack, and SECA. You will walk away understanding exactly how we can unite this fragmented landscape through shared values and common open standards , and why your specific role as an open-source donator, enterprise adopter, or individual contributor is crucial to driving the success of the cloud native ecosystem in Europe.

Speaker: Daniel Sterz
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="0" average="0.00" min="1" max="5"/>
    <media:statistics views="29"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:byehnUkET6I</id>
  <yt:videoId>byehnUkET6I</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>Designing Organizations to Prevent Incidents - Adelina Stanciu</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=byehnUkET6I"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:44+00:00</published>
  <updated>2026-10-15T04:12:05+00:00</updated>
  <media:group>
   <media:title>Designing Organizations to Prevent Incidents - Adelina Stanciu</media:title>
   <media:content url="https://www.youtube.com/v/byehnUkET6I?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i3.ytimg.com/vi/byehnUkET6I/hqdefault.jpg" width="480" height="360"/>
   <media:description>This session explores how organizations can prevent production incidents by design, rather than relying on reactive fixes after failures occur.

In cloud-native environments, incidents are rarely caused by a single technical failure. They often emerge from the interaction between systems, teams, and operational practices. This talk examines how organizational structures, workflows and decision-making models directly influence security, reliability and compliance in production.

Using real-world examples, we'll analyze how organizations reduce incident risk by designing teams around clear service ownership, aligning ownership with authority, creating early risk feedback loops, improving cross-team coordination and designing for reliable recovery.

The session shows how organizational design shapes production outcomes and how intentional structures lead to fewer incidents and more predictable, low-disruption operations.

Speaker: Adelina Stanciu
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="1" average="5.00" min="1" max="5"/>
    <media:statistics views="38"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:j1kyiO27r0s</id>
  <yt:videoId>j1kyiO27r0s</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>Don’t Debug, Reset: Managing Kubernetes the Declarative Way with Talos Linux - Daniel Bodky</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=j1kyiO27r0s"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:44+00:00</published>
  <updated>2026-10-15T05:06:02+00:00</updated>
  <media:group>
   <media:title>Don’t Debug, Reset: Managing Kubernetes the Declarative Way with Talos Linux - Daniel Bodky</media:title>
   <media:content url="https://www.youtube.com/v/j1kyiO27r0s?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i3.ytimg.com/vi/j1kyiO27r0s/hqdefault.jpg" width="480" height="360"/>
   <media:description>Self-hosting Kubernetes means owning everything cloud providers happily hide from you. Configuration, maintenance, and security burdens compound with each layer of the stack. I used to reach for kubeadm or Ansible playbooks and ended up with a collection of imperative commands that made sense six months ago, but not at 1am on a Saturday.

Talos Linux takes a different approach with a minimal attack surface, API-driven node and cluster management, and secure defaults. When nodes go dark, I reset instead of debug.

Through live demos drawn from my personal homelab, we will walk through typical cluster administration tasks and see how Talos handles them in ways that are scalable and reproducible.

You will leave with a solid mental model for why a minimal distro paired with immutability is a superpower and not a constraint, and a practical path to apply Talos whether your cluster lives on ARM boards on your desk, in cloud VMs, or at the edge.

Speaker: Daniel Bodky
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="2" average="5.00" min="1" max="5"/>
    <media:statistics views="120"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:I3XqMW0jOB0</id>
  <yt:videoId>I3XqMW0jOB0</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>How We Run 8,000+ GPUs on Kubernetes with Slurm - Fagani Hajizada &amp; Giulio Calzolari</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=I3XqMW0jOB0"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:42+00:00</published>
  <updated>2026-10-03T05:52:01+00:00</updated>
  <media:group>
   <media:title>How We Run 8,000+ GPUs on Kubernetes with Slurm - Fagani Hajizada &amp; Giulio Calzolari</media:title>
   <media:content url="https://www.youtube.com/v/I3XqMW0jOB0?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i2.ytimg.com/vi/I3XqMW0jOB0/hqdefault.jpg" width="480" height="360"/>
   <media:description>As AI workloads scale to thousands of GPUs, organizations need scheduling capabilities that Kubernetes doesn't provide out of the box: gang scheduling for multi-node jobs, fair-share policies across teams, and topology-aware placement for optimal GPU communication. Slurm has solved these problems for decades in HPC, but traditionally requires dedicated bare-metal infrastructure.

In this talk, we show how to run Slurm natively on Kubernetes using Slinky, an open-source operator. This approach lets you keep existing Slurm workflows (job scripts, accounting, fair-share policies) while gaining Kubernetes benefits: declarative configuration, rolling updates, Helm deployments, and ecosystem integration. We operate this at scale with 8,000+ GPUs for distributed AI training.

We'll share our journey deploying and operating Slurm on Kubernetes, covering the obstacles we hit, how we solved them, and what we learned. You'll learn how to deploy Slurm on your existing Kubernetes cluster, how to keep both systems in sync when nodes fail or need maintenance, and common pitfalls to avoid. This approach works whether you have 8 GPUs or 8,000, and you can try it on your own cluster today.

How will your presentation benefit the ecosystem?

Our teams have years of experience with Slurm, and the job scripts, fair-share policies, and accounting workflows are well established across the HPC community. Attendees will learn how to run these workloads on Kubernetes without rebuilding infrastructure.

Slinky is fully open source, and every feature we discuss is available in the upstream project. This talk bridges the HPC and cloud-native communities by showing a practical path to combine both worlds.

Attendees will leave knowing whether Slurm-on-Kubernetes fits their use case, how to get started with a minimal setup, and what challenges to expect.

Speaker: Fagani Hajizada &amp; Giulio Calzolari
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="14" average="5.00" min="1" max="5"/>
    <media:statistics views="534"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:JrAOHnJ6fn0</id>
  <yt:videoId>JrAOHnJ6fn0</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>Build Your Own Cloud Native DBaaS: Leveraging GitOps and CNPG - Shiva deep Gundoju</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=JrAOHnJ6fn0"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:42+00:00</published>
  <updated>2026-10-15T05:05:42+00:00</updated>
  <media:group>
   <media:title>Build Your Own Cloud Native DBaaS: Leveraging GitOps and CNPG - Shiva deep Gundoju</media:title>
   <media:content url="https://www.youtube.com/v/JrAOHnJ6fn0?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i3.ytimg.com/vi/JrAOHnJ6fn0/hqdefault.jpg" width="480" height="360"/>
   <media:description>In this session, we move beyond the &quot;stateless-only&quot; mindset to explore how you can build a self-service Database-as-a-Service (DBaaS) platform.
During this session, the concept of a self-managed database platform is practically demoed using a live prototype. We will walk through a local-first setup powered by Kind, Harness CI/CD, Argo CD, and CNPG. This demonstration provides a hands-on look at how to get started with this approach and how to move in a direction that successfully bridges the gap between a local environment and a production-ready GitOps workflow.
Many teams are still afraid to run stateful workloads on Kubernetes. Databases, in particular, are often viewed as a &quot;high-wire act&quot; because of perceived complexity and the multiple layers of management they require. However, many organizations are already reaping the benefits of stateful Kubernetes by configuring and managing these systems the right way. Cloud-native operators, specifically CloudNativePG (CNPG), have truly changed the game, making it significantly easier to manage high-availability Postgres clusters through simple, declarative YAML.

What you will learn:
The Operator Advantage: How CloudNativePG simplifies complex tasks like automated failover, backups, and WAL archiving.
GitOps for Databases: Using Harness and Argo CD to treat your data infrastructure as code, ensuring your clusters stay in sync with your Git repository.
Local-First Prototype: How to simulate a production-grade DB environment on your laptop using Kind and Docker Desktop.
Why attend?
If you have been hesitant to move your databases into Kubernetes, this session provides a clear roadmap. You will see a functional prototype that proves stateful workloads don't have to be scary when you have the right automation and operators in place.

Speaker: Shiva Deep Gundoju
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="0" average="0.00" min="1" max="5"/>
    <media:statistics views="110"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:eWBGPnGhKLY</id>
  <yt:videoId>eWBGPnGhKLY</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>The Limits of Vibe Coding: AI, Hype, and the Future of Real Software Engineering - Christian Gläser</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=eWBGPnGhKLY"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:42+00:00</published>
  <updated>2026-10-15T04:36:25+00:00</updated>
  <media:group>
   <media:title>The Limits of Vibe Coding: AI, Hype, and the Future of Real Software Engineering - Christian Gläser</media:title>
   <media:content url="https://www.youtube.com/v/eWBGPnGhKLY?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i2.ytimg.com/vi/eWBGPnGhKLY/hqdefault.jpg" width="480" height="360"/>
   <media:description>AI coding assistants can generate working code in seconds—but production systems require much more than code that compiles. This has led to a new development style sometimes called “vibe coding”—iteratively prompting AI until the output appears to work.

While this approach can be useful for experimentation and rapid prototyping, production systems require much more than code that compiles. They require domain knowledge, architectural thinking, structured data modeling, and the ability to critically evaluate generated output.

Modern AI tools are extremely good at producing large volumes of plausible code. The challenge is that this code often looks correct while quietly introducing architectural issues, hidden complexity, and long-term maintenance problems. Teams may find themselves maintaining significantly more code than before—without necessarily improving system quality.

This talk examines the current AI wave through the lens of the Gartner Hype Cycle and explores where AI tools genuinely improve developer productivity—and where human expertise remains essential.

A key part of this shift is the emergence of prompt engineering as a design discipline. AI tools work best when developers follow a design-first approach: defining system architecture, data models, and constraints before asking AI to generate implementation details.

In this model, developers increasingly act as orchestrators of AI tools, delegating tasks such as boilerplate generation, testing, documentation, and refactoring to a “virtual team” of assistants. But this also requires the ability to structure problems clearly and detect when AI-generated code is incorrect or harmful for long-term maintainability.

The key insight: AI does not remove the need for expertise—it amplifies it.

Good engineers can use AI to accelerate development dramatically. Without strong architectural thinking and critical oversight, however, AI can just as quickly generate large amounts of technical debt.

This session provides a pragmatic perspective on how engineers can use AI responsibly while maintaining high-quality, maintainable cloud-native systems.

Speaker: Christian Gläser
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="0" average="0.00" min="1" max="5"/>
    <media:statistics views="24"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:fNMNj96JItQ</id>
  <yt:videoId>fNMNj96JItQ</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>Your Backstage, Your Problems, Your Metrics - Thomas Schuetz &amp; Katharina Sick</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=fNMNj96JItQ"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:39+00:00</published>
  <updated>2026-10-15T04:37:38+00:00</updated>
  <media:group>
   <media:title>Your Backstage, Your Problems, Your Metrics - Thomas Schuetz &amp; Katharina Sick</media:title>
   <media:content url="https://www.youtube.com/v/fNMNj96JItQ?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i3.ytimg.com/vi/fNMNj96JItQ/hqdefault.jpg" width="480" height="360"/>
   <media:description>Backstage is often proposed as the solution to all platform problems. Just deploy it and watch the magic happen. Too good to be true? Exactly. Let's take a step back: which problems are you actually trying to solve? Service discovery? Onboarding? Tool sprawl? Documentation chaos? Your problems are unique. Just as the metrics for your developer portal should be.

A developer portal without a purpose is only just another piece of software, causing toil and maintenance effort. In this talk, we'll show you how to apply a product mindset to your platform (and Backstage): starting with identifying the specific problems you want to solve before implementing Backstage, define metrics for those problems, and measure whether Backstage actually delivers value and how it can be improved. Think DORA metrics, but for your internal developer portal.

We’ll equip you with a framework to validate your platform decisions with data, whether you're considering Backstage or already running it in production.

Speaker: Thomas Schuetz &amp; Katharina Sick
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="0" average="0.00" min="1" max="5"/>
    <media:statistics views="30"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:sRbaMwAb3ps</id>
  <yt:videoId>sRbaMwAb3ps</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>The AI-Empowered Team: A PM's Guide to What Actually Works - Dominik Schmidle</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=sRbaMwAb3ps"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:39+00:00</published>
  <updated>2026-10-15T03:45:55+00:00</updated>
  <media:group>
   <media:title>The AI-Empowered Team: A PM's Guide to What Actually Works - Dominik Schmidle</media:title>
   <media:content url="https://www.youtube.com/v/sRbaMwAb3ps?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i4.ytimg.com/vi/sRbaMwAb3ps/hqdefault.jpg" width="480" height="360"/>
   <media:description>Over the past year, our platform teams at Giant Swarm have been running a real experiment: hand over the boring stuff to AI agents and see what happens. Ticket writing, documentation, repetitive formatting - the tasks that eat hours but don't need strategic thinking.

Some things failed spectacularly. Others genuinely work. The key insight: better context beats better prompts, every time. Treat AI like a junior team member who needs training. Build and refine evals to measure actual quality, not vibes. And know when a simple script beats an agent.

This talk shares what stuck: practical patterns for AI-augmented development work, real examples of what flopped and why, and honest expectations about what AI can actually do for your team today. For PMs, tech leads, product teams and anyone curious about AI beyond the hype.

Speaker: Dominik Schmidle
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="0" average="0.00" min="1" max="5"/>
    <media:statistics views="33"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:Ko2IIWk5RPE</id>
  <yt:videoId>Ko2IIWk5RPE</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>The Evolution of GitOps in Platform Engineering - Koray Oksay &amp; Artem Lajko</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=Ko2IIWk5RPE"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:38+00:00</published>
  <updated>2026-10-15T05:07:41+00:00</updated>
  <media:group>
   <media:title>The Evolution of GitOps in Platform Engineering - Koray Oksay &amp; Artem Lajko</media:title>
   <media:content url="https://www.youtube.com/v/Ko2IIWk5RPE?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i4.ytimg.com/vi/Ko2IIWk5RPE/hqdefault.jpg" width="480" height="360"/>
   <media:description>GitOps in 2017 as a simple model for application delivery, built around four core principles. Today, it has evolved into a central part of platform engineering, where teams operate large numbers of clusters and shared platform services.

This talk shows how GitOps has evolved over time, why early approaches break at scale, and how everything-as-code enabled automation while also creating configuration sprawl. Tools like Argo CD, Helm, and Kustomize made scaling possible, but often hide the real state of a system.

We will look at the limits of Git as a state store and why file-based, pull-driven reconciliation does not scale well. This led to OCI-based delivery, Gitless GitOps, and new approaches that aim to provide a real, operable state store.

Finally, we will explain why this matters now. Platform teams need visibility, regulation like the Cyber Resilience Act requires a clear infrastructure supply chain, and AI-driven operations depend on access to the final, real manifests.

Speaker: Artem Lajko &amp; Koray Oksay
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="13" average="5.00" min="1" max="5"/>
    <media:statistics views="847"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:T5gfXqeZufA</id>
  <yt:videoId>T5gfXqeZufA</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>Keynote: The Evolution to Platform Engineering 2.0 - Bjoern Brundert</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=T5gfXqeZufA"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:37+00:00</published>
  <updated>2026-10-15T04:25:28+00:00</updated>
  <media:group>
   <media:title>Keynote: The Evolution to Platform Engineering 2.0 - Bjoern Brundert</media:title>
   <media:content url="https://www.youtube.com/v/T5gfXqeZufA?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i1.ytimg.com/vi/T5gfXqeZufA/hqdefault.jpg" width="480" height="360"/>
   <media:description>Platform Engineering 1.0 succeeded in reducing developer cognitive load but is now constrained by rigid architectures, a developer-only focus, and reactive security and FinOps. New forces demand a strategic evolution. This session outlines &quot;Platform Engineering 2.0&quot;: an AI-native, autonomous, and business-aligned framework built on five core pillars. Discover how to implement the Multi-Persona Experience, Embedded FinOps, Security Shifts Down, More Composable Architecture, and an AI Native Platform to future-proof your organization. You will leave with strategic recommendations for executives and platform engineers to mandate a PE 2.0 business case, reframing the platform as strategic infrastructure tied to developer velocity, cloud cost reduction, and AI readiness metrics.

Speaker: Bjoern Brundert
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="0" average="0.00" min="1" max="5"/>
    <media:statistics views="33"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:Zqrc1zrg9uQ</id>
  <yt:videoId>Zqrc1zrg9uQ</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>eBPF on Wheels - Container Security for Automotive Use Cases - Reinhard Kugler</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=Zqrc1zrg9uQ"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:32+00:00</published>
  <updated>2026-10-15T04:58:00+00:00</updated>
  <media:group>
   <media:title>eBPF on Wheels - Container Security for Automotive Use Cases - Reinhard Kugler</media:title>
   <media:content url="https://www.youtube.com/v/Zqrc1zrg9uQ?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i3.ytimg.com/vi/Zqrc1zrg9uQ/hqdefault.jpg" width="480" height="360"/>
   <media:description>Automotive and industrial use cases share a common problem: updateability and observability in constrained and hazard-tested environments. Container technologies are known as successful deployment patterns in the Cloud and have been adopted to the automotive and industrial domain. This should provide continuous updates and flexible deployment, but how is the state with attack detection in those domains?

The use of physical interfaces and non-IP bus systems, such as the CAN bus or on-board interfaces are uncommon in the IT world and therefore new territory for cloud-native. As Red Hat and Canonical entered the road of edge devices, the Linux platform quickly matured in safety-critical domains and paved the way. The missing piece is is the deep protocol inspection in electronic systems.

This talk gives an overview of how containers and eBPF can be employed in automotive to detect cyber attacks.

Speaker: Reinhard Kugler
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="1" average="5.00" min="1" max="5"/>
    <media:statistics views="22"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:r4tro8X_wy0</id>
  <yt:videoId>r4tro8X_wy0</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>Sovereignty through Mastery: Building a Multi-Cloud IDP for Digital Autonomy - Andreas Grub</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=r4tro8X_wy0"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:31+00:00</published>
  <updated>2026-10-15T04:27:39+00:00</updated>
  <media:group>
   <media:title>Sovereignty through Mastery: Building a Multi-Cloud IDP for Digital Autonomy - Andreas Grub</media:title>
   <media:content url="https://www.youtube.com/v/r4tro8X_wy0?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i3.ytimg.com/vi/r4tro8X_wy0/hqdefault.jpg" width="480" height="360"/>
   <media:description>With the shifting geopolitical landscape, sovereignty has moved from a legal requirement to a core business resilience strategy. However, true sovereignty isn’t achieved by isolation, but through the freedom of choice. In this session, we demonstrate how an Internal Developer Platform (IDP) serves as the operational backbone for this autonomy. We will deep-dive into a production-ready stack that bridges the gap between global hyperscalers and European sovereign clouds, exploring the architectural decisions needed to move beyond fragmented infrastructure and become sovereign by design:

- The Unified Experience: How to bring separate cloud teams and varying levels of maturity into a single interface, making sovereign options as easy to discover and use as global ones.
- Designing for Developer Appeal: Why sovereignty fails if it’s a burden. We’ll discuss how to provide &quot;meaningful choice&quot; by creating abstractions that make local providers as developer-friendly as hyperscalers.
- Mastery Over Magic: Why we move away from the myth of &quot;automatic workload migration&quot; and focus instead on building the internal capability and technical standards to switch providers whenever the business demands it.

Speaker: Andreas Grub
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="0" average="0.00" min="1" max="5"/>
    <media:statistics views="16"/>
   </media:community>
  </media:group>
 </entry>
 <entry>
  <id>yt:video:0RmPD7stN4M</id>
  <yt:videoId>0RmPD7stN4M</yt:videoId>
  <yt:channelId>UCdhKM1VoQLhP261oSm79LKA</yt:channelId>
  <title>Don't Just Flip the Switch: Measure What Happens Next - Lukas Reining &amp; André Silva</title>
  <link rel="alternate" href="https://www.youtube.com/watch?v=0RmPD7stN4M"/>
  <author>
   <name>Cloud Native Summit Munich</name>
   <uri>https://www.youtube.com/channel/UCdhKM1VoQLhP261oSm79LKA</uri>
  </author>
  <published>2026-07-19T20:37:20+00:00</published>
  <updated>2026-10-15T04:30:09+00:00</updated>
  <media:group>
   <media:title>Don't Just Flip the Switch: Measure What Happens Next - Lukas Reining &amp; André Silva</media:title>
   <media:content url="https://www.youtube.com/v/0RmPD7stN4M?version=3" type="application/x-shockwave-flash" width="640" height="390"/>
   <media:thumbnail url="https://i1.ytimg.com/vi/0RmPD7stN4M/hqdefault.jpg" width="480" height="360"/>
   <media:description>You're gradually rolling out a new feature to your users. But what's actually happening? Are users in region A seeing faster checkouts while region B is quietly timing out? Is version B of your pricing page converting better, or are error rates creeping up in ways your dashboard doesn't show? Most teams can tell you a flag is on. Fewer can tell you what it's actually doing to their system and users.

OpenFeature gives you a vendor-agnostic way to dynamic feature flag evaluation. OpenTelemetry gives you vendor-agnostic observability. Both are CNCF projects, and when you wire them together, flag evaluations show up in your traces, metrics, and events. You can go from &quot;it's live&quot; to &quot;here's the P95 latency impact per region, per variant, per user cohort&quot;.

This talk walks through that setup end to end: instrument an app, roll out a feature, then use the telemetry to understand the real impact. We'll also get into traces vs. metrics vs. logs. They each have a sweet spot, and picking wrong costs you either in cardinality or in blind spots.

Speaker: Lukas Reining &amp; Andre Silva
Slides: 

Find out more on https://cloudnativesummit.de/ 
Follow us on: 
LinkedIn - https://www.linkedin.com/company/cns-munich Twitter/X - https://x.com/cnsmunich 
Instagram - https://www.instagram.com/cnsmunich 
BlueSky - https://bsky.app/profile/cnsmunich.bsky.social</media:description>
   <media:community>
    <media:starRating count="1" average="5.00" min="1" max="5"/>
    <media:statistics views="17"/>
   </media:community>
  </media:group>
 </entry>
</feed>
