A real-world-inspired architecture talk about embedding an AI agent into the operational workflow of a highload data pipeline. We walk through a cascade failure scenario: corrupted data enters the pipeline, Kafka queues get stuck, storage pressure grows, thousands of Kubernetes pods start failing and rescheduling, etcd degrades, and PostgreSQL becomes a secondary pressure point. Then we show how an agent built with AWS Bedrock AgentCore, LangChain, and MCP/Gateway could detect early signals, isolate corrupted messages, suggest human-approved fixes, protect cluster stability, and turn noisy telemetry into actionable recovery steps.

In this presentation, we will explore a practical use case of implementing effective infrastructure autoscaling using HPA, VPA, and Cluster Autoscaler. While working with standard VPA, we encountered several limitations, including a lack of flexibility in configuring calculation intervals and conflicts when running concurrently with HPA. Consequently, we decided to develop our own custom VPA controller. In our new solution, we: - Achieved stable coexistence of VPA and HPA on the same resources. - Implemented a filtering mechanism for transient CPU spikes during the pod startup phase. - Optimized the architecture by consolidating the functionality of three standard components into a single pod. - Leveraged the new In-Place Pod Resize capabilities introduced in Kubernetes 1.33. Key result: Optimized resource consumption and a 20–40% reduction in infrastructure costs.

This is a story about real pain and the maturation of a logging system. We’ll examine how the lack of standards breaks observability and why Kubernetes became the point of no return, forcing us to rethink our entire logging approach. We’ll walk through the requirements and architectural decisions that helped us regain control. I’ll share hands-on experience in building controlled, production-grade log pipelines - without magic and without “silver bullet” tools. This is an honest story from real production environments.

This talk is dedicated to the practical experience of transitioning from the classic Istio architecture with sidecar containers to a new model - Istio Ambient Mesh. We will tell how scaling a microservice infrastructure on Kubernetes led to increased resource costs, complicated CI/CD, and delays in launching services. The search for solutions led us to implement Ambient Mesh, an architecture without sidecar containers that greatly simplifies mesh integration and reduces infrastructure costs. The presentation will detail the technical aspects of the migration: how we prepared for the transition, what challenges we faced during the implementation process, and how we managed to overcome them in cooperation with the Istio community. We will share the results, analytics, real before/after metrics, and give practical advice to teams planning to implement Ambient Mesh in production.

When human safety is at stake, technical reliability is not just a requirement. In this talk, we'll look at architecture, workloads, WebSocket solutions, Kubernetes scaling, and other technical aspects of creating an airborne alarm map. This is a story not only about code, but also about responsibility.

Kubernetes Controllers and Operators are a trending topic in conferences, interviews, and production today. I will share the story of the evolution of our Promotion (Release) system, from simple Kubernetes API REST calls to Informers and Controllers, based on my own experience. This story is particularly interesting because it serves as a great case for personal growth for you as well as for your DevOps/SRE team. It touches on Kubernetes architecture details, Networking, GitOps, IaC, Caching, development patterns, and Golang data structures. Even if you have no development experience (as is the case for most of our team), I will share how a Cursor AI assistant became yet another — though virtual — engineer on our team. Bonus: 10 years of Kubernetes & trends KubeCon24 North America.

There are many opinions regarding the deployment of databases in Kubernetes. However, the strong views of random individuals may not assist you in making the right decision for your specific situation. In this talk, you will receive a comprehensive list of problems and questions to address before moving your production database into Kubernetes. With that list, you will gain a better understanding of the associated risks, prepare for potential failures, and make an informed decision.

At the conference, I will present an approach to extending the capabilities of Kubernetes, transforming it into a fully-fledged private Cloud with support for virtualization, isolated networks, and multi-organization authentication.

There are many opinions regarding the deployment of databases in Kubernetes. However, the strong views of random individuals may not assist you in making the right decision for your specific situation. In this talk, you will receive a comprehensive list of problems and questions to address before moving your production database into Kubernetes. With that list, you will gain a better understanding of the associated risks, prepare for potential failures, and make an informed decision.

Kubernetes Controllers and Operators are a trending topic in conferences, interviews, and production today. I will share the story of the evolution of our Promotion (Release) system, from simple Kubernetes API REST calls to Informers and Controllers, based on my own experience. This story is particularly interesting because it serves as a great case for personal growth for you as well as for your DevOps/SRE team. It touches on Kubernetes architecture details, Networking, GitOps, IaC, Caching, development patterns, and Golang data structures. Even if you have no development experience (as is the case for most of our team), I will share how a Cursor AI assistant became yet another — though virtual — engineer on our team. Bonus: 10 years of Kubernetes & trends KubeCon24 North America.