We’ll explore how modern teams are building internal AI platforms, augmenting Kubernetes with intelligent agents, and running large-scale inference on EKS.

From real-world architectures to open-source tools like Kagent and KServe, discover what “AI Platform Engineering” really means in 2025, and how to prepare your stack for it.

Oh, and yes, there’ll be good food, cold drinks, and plenty of hallway conversations. 🍕🍺

🕔 17:30 – 18:00 Welcome, Mingling, Food & Drinks

Kick off the evening with great vibes! Grab a bite, meet fellow cloud-native and AI enthusiasts, and chat with engineers, architects, and innovators shaping the future of platform engineering.

🕕 18:00 – 18:30 What is AI Platform Engineering and Why Should You Care

Speaker: Engin Diri, Senior Solutions Architect at Pulumi

In 2025, generative AI is no longer an experiment, it’s part of production systems, and platform engineering teams are under pressure.

How can they let developers move fast with AI while still holding onto the hard-won practices of good platform engineering?

This talk explores that tension through real experiences building internal AI platforms. Starting with the familiar foundations - Kubernetes, containers, and automation - it expands into the new demands of GenAI-Ops, including RAG, fine-tuning workflows, and autonomous agent orchestration.

You’ll leave with a clearer picture of what AI platform engineering truly means, why it’s not just rebranded data science, and how to position your platform team at the center of your company’s AI strategy.

About Engin Diri - Senior Solutions Architect at Pulumi, with over 15 years of experience in IT across the Schwarz Group and W&W Versicherungen. Engin brings deep expertise in enterprise systems and modern platform architectures, now helping organizations bridge the gap between developers and infrastructure in the age of AI.

🕡 18:30 – 19:00 Augment Your Kubernetes with AI Agents Using Kagent

Speaker: Ant(on) Weiss, Software Delivery Futurist at PerfectScale by DoiT

Kubernetes can be a powerful platform for building and running AI agents using Kagent, an open-source framework for DevOps and platform engineers to automate complex operations and troubleshoot workloads.

In this hands-on session, we’ll deploy Kagent in lab clusters, create AI agents, and enrich them with custom tools. You’ll see how agents can manage and troubleshoot Kubernetes workloads, and get a taste of self-hosting LLMs using Ollama or Kaito, including discussions on cost and performance trade-offs.

It’s time to make your clusters agentic!

About Ant(on) Weiss - With over 20 years in tech, spanning engineering, marketing, and leadership roles, Ant(on) focuses on software delivery optimization and human collaboration. A sought-after speaker, writer, and trainer, he blends expertise in DevOps, Lean, Continuous Delivery, and Cloud Native Systems with thought-provoking storytelling.

🕖 19:00 – 19:30 High-scale Inference with EKS and KServe

Speaker: DevOps Team, Gong

Learn from how to master AI workload efficiency on Amazon EKS through Gong's practical journey of optimization. This technical session unveils their proven strategies across three critical areas: implementing serverless serving with KServe/KNative to eliminate cold starts, maximizing GPU cost savings through innovative sharing techniques and Spot instance usage, and building robust AI observability using DCGM and OCI image caching. Join us to discover how Gong.io transformed its AI infrastructure to achieve both performance and cost optimization, and leave with actionable insights you can apply to your own EKS environment.