Abstract

Kubernetes is easy, isn't it? Creating Kubernetes cluster in public cloud take a few minutes, deploying an application it's like same. But how can we make sure that our cluster scales in an efficient way? And what can we do with workloads which are not really a good fit for Kubernetes?

Bio

Kim is Regional Field CTO and Solution Architect at CAST AI, focused on guiding Kubernetes users to cost efficient and automated platforms. Before CAST, he was working as Cloud Architect and Consultant for Cloudical, noris network and Deutsche Telekom. He’s passionate about cloud infrastructure, automation and open source technologies. He’s an open source contributor and public speaker.

Abstract

If you ask 10 DBAs in a conference about putting data on Kubernetes, most will say that’s a bad idea – Divine. Divine is an advocate for data on Kubernetes, and a Data on Kubernetes Ambassador. Shivadeep, on the other hand, as an Oracle DBA, traditionally believed that containers and Kubernetes weren't suitable for persistent workloads like databases. And that’s rightly so because it was originally designed to be stateless. But he didn’t know a lot had changed. With no prior experience in containers, Kubernetes, or NodeJS, Shivadeep embarked on a challenging project to build a scalable Pacman game. This journey not only helped him learn new technologies but also changed his perspective on using Kubernetes for database workloads. Shivadeep discovered why and how Kubernetes is increasingly becoming a popular choice for running databases as well. Shivadeep will start this talk sharing his story of learning and exploration, highlighting: - How he containerized and deployed the PACMAN Game application on Kubernetes - How he connected it to an Oracle DB on Container as well as on-prem Oracle database via CMAN To make the session even more engaging, Shivadeep will provide a hands-on opportunity for attendees to try out the Pacman game! Towards the end of the talk, Divine will share how DBAs can evolve to manage databases on Kubernetes.

Bio

Currently the Founder of EverythingDevOps, Divine Odazie is a Developer Relations Professional and Technical Writer with over 5+ years of experience in Technology and a track record in Backend Engineering, DevOps, Cloud Native and Developer Relations on a global scale. He has given talks/workshops at developer conferences like Open Source Summit Europe, KubeCon North America, Cloud Native Rejekts, and Ansible Contributor Summit. He also supports the cloud native community in Africa as a KCD (Kubernetes Community Days) Organizer, and other communities empowering Africa with Technology.

Shivadeep Gundoju is a seasoned database administrator at ING Netherlands , who loves to automate everything around databases and its Infrastructure. He holds a Master's degree in Computing Systems & Infrastructure and excited to explore Oracle and other RDBMS on containers/Kubernetes.

Abstract

Instrumenting legacy or closed sourced applications can be a pain. But it doesn't have to be! This talk offers an introduction to the open source Beyla tool and shows you how to utilize it in order to instrument different types of applications without touching a single line of code. Beyla hooks into the kernel using eBPF and does the hard work for you. It's not only useful for services where you can't influence the underlying instrumentation, but it can help you establish a baseline level of observability in large scale deployments. We’ll also go over limitations, so you’ll leave the talk knowing if Beyla is right for you (spoiler: it probably is).

Bio

Dominik started his journey in technology as an SRE, working on projects ranging from warehouse logistics and photobook designers to analyzing satellite imagery. During this time, he discovered his passion for developer tooling and making sure developers can focus on what they do best - build great software! Now he is working as a Developer Experience Engineer at Grafana Labs, building tools to see clearly in the ever-changing world of software.

Abstract

Watch AI agents autonomously solve production crises! Azure SRE Agent and Hawkeye collaborate to investigate a critical error, diving into Azure and multi-cloud telemetry. See real-time coordination as agents leverage source code to deploy fixes, demonstrating the future of cross-cloud SRE.

Bio

Dheeraj Bandaru is a Product Manager at Microsoft Azure working on the Azure SRE Agent initiative—pioneering one of the industry's first production-ready autonomous SRE platform—and Azure Static Web Apps, where he has driven adoption across thousands of customers. His work spans the full Azure PaaS portfolio including Azure App Service, Azure Container Apps, and Azure Functions.

Francois Martel is a Field CTO at Neubird.ai, where he helps enterprise SRE teams transform their operations through Generative AI. With over two decades of experience in cloud architecture and AI/ML, François specializes in addressing fundamental challenges in modern IT operations through Hawkeye, a GenAI-powered SRE solution that automates routine investigations and enables engineering teams to focus on strategic innovation.

Previously at Amazon, François led Analytics and AI/ML initiatives for Financial Services, developing comprehensive GenAI programs and driving enterprise-scale transformations. He brings deep technical expertise in cloud-native architecture, machine learning, and distributed systems, holding certifications in AWS Solution Architecture, Analytics, ML, and Generative AI, along with Kubernetes administration.

Connect with François to discuss the future of SRE, GenAI's impact on IT operations, and building resilient engineering teams."

Abstract

Are your observability signals trapped in separate pillars? Logs in one place, metrics in another, both losing context? At ClickHouse, we faced this challenge at a massive scale. Our solution was to abandon the traditional model and embrace a new philosophy: store everything, aggregate nothing. This talk charts our journey to 100 PB and 500 trillion rows, centered on the concept of "wide events." Instead of shipping a simple log message and a separate, pre-aggregated metric, we store a single, context-rich event containing every possible dimension. This shift from three pillars to a single warehouse of high-cardinality telemetry was a game-changer. The key to this model is using ClickHouse itself as our observability backend. This unlocks unbounded query flexibility through full SQL. When an engineer asks "what's the p95 pod replacement time after termination?", we don't say "let me ship a new metric." We write a SQL query. This talk will cover: The Wide-Event Philosophy: Why logging rich, structured events is more powerful than juggling separate logs and metrics, and how it defeats cardinality fears. Unbounded SQL-based Querying: We’ll show real-world examples of complex diagnostic queries (like using ASOF JOIN to correlate disparate Kubernetes events) that are impossible in traditional log-search tools. The Enabler - SysEx: How our custom, high-performance exporter made this firehose of wide-event data technically and financially viable, handling 20x the volume of OTel with 90% less CPU. Beyond Dashboards: How this approach allows us to use data science tools like Plotly and Jupyter directly on our observability data for deeper, more flexible analysis. Join us to see how treating observability as a data warehouse problem, powered by ClickHouse, gives you the speed and flexibility to answer any question about your systems, past or present.

Bio

Rory is a senior engineer on the observability team at ClickHouse, responsible for building and managing a 100+ petabyte observability platform. With over a decade of SRE experience, he focuses on monitoring ClickHouse Cloud, optimizing metrics pipelines and working with Grafana for visualization.

Abstract

Air-gapped systems are seen as the pinnacle of security, but are they truly untouchable? This talk explores real-world breaches—from Stuxnet to electromagnetic attacks—highlighting modern threats like supply chain risks and social engineering. Attendees will learn practical strategies to strengthen air-gapped environments through physical security, procedural controls, and advanced detection methods.

Bio

Sean Behan is a Senior Offensive Security Engineer at Oracle with a decade of experience across top-tier organizations including Google, AWS, the NSA, and the U.S. Navy. He specializes in offensive security, red teaming, and cloud security, with deep technical expertise in exploit development and adversary simulation. Sean holds OSCP and CISSP certifications and is passionate about continuous learning, regularly participating in CTFs and advanced cybersecurity research.

Abstract

Managing alerts across hundreds of services can quickly become a challenge — fragmented workflow, inconsistent configurations and high operational costs. In this session, we will share how we streamlined alerting by adopting Kibana Alerts and building a bidirectional automation on top of it, using a GitOps-driven approach. Our solution balances UI flexibility with Infrastructure as Code (IaC) principles, simplifying configuration while maintaining control and simplifying complex features at scale. Discover how automation improved consistency, enabled version control, and reduced complexity.

In this session, we’ll explore how we automated Kibana alerting at scale—and tackled some of the gaps that come with it. You’ll see how we implemented advanced features like re-alerting and spike detection, even though they weren’t available out of the box. From GitOps-driven pipelines to creative workarounds, this talk is a practical deep dive into making alerting reliable, repeatable, and ready for scale.

Bio

Ayd is a DevOps Team Lead at AUTO1 Group, where he helps power one of Europe’s fastest-growing digital marketplaces by building scalable, resilient, and automated infrastructure. With over 12 years of experience across cloud, DevOps, and platform engineering, he’s led transformations from legacy chaos to modern observability at scale. Known for breaking things on purpose (and then automating the fix), Ayd is passionate about turning operational complexity into repeatable systems. Outside the terminal, he’s a TEDx organizer, TED Translator, and self-declared youth activist. He believes in sharing what hurts, building what works, and always pushing platforms—gently or otherwise—toward what they should become.

Abstract

Platform engineering must embrace resilience, not just scale. This talk shows how Kubernetes, ArgoCD, Chaos Mesh, and Pixie, using eBPF, enable real-time observability, chaos testing, and automated deployments, making platforms stable, chaos-ready, and easy to manage.

Bio

10 Years of experience in the world of Cloud and Automation. Scott is the lead architect in the CTO office at TeraSky and leads the platform engineering and cloud native initiatives. Scott has vast experience with both the legacy datacenter and cutting-edge Public Cloud and cloud native technologies. He specializes in bridging the gap for customers and easing their transition to the new digital transformation era. Scott is active in the Kubernetes project and its ecosystem tools, including Pinniped, CusterAPI, Carvel, Cartographer, and KubeApps.

Abstract

The shift from traditional testing to chaos engineering marks a revolution in building reliable systems. This session unpacks the concept’s history and role in ensuring system resilience. We’ll look at some of the approaches to chaos engineering, before looking at Chaos Engineering as a Service with Amazon’s Fault Injection Service. You’ll leave with insights into crafting and running Fault Injection Service experiment templates, plus a live demo looking at how we can now test serverless code in AWS with Chaos engineering.

Bio

Starting out in a very traditional background in the data-centres of the fabled M4 corridor, Simon eventually realised it was time to give up on the problems of manually babysitting servers, racks and UPS's and migrated to the Cloud (and in the process, the Scottish Highlands) and now works to enable clients to transition their workloads, processes and swag requirements to the same route. Simon is a member of the AWS Community builder program, and these days enjoys coaching and mentoring as much getting hands-on with code, automation and head-scratching.

Abstract

As AI systems scale in complexity and impact, observability becomes essential—not just for performance monitoring, but for ensuring quality, safety, and trust. In Azure AI Foundry, evaluators are the backbone of this observability layer, enabling continuous, automated assessments across the AI lifecycle. This session provides a deep technical dive into the evaluator framework in Azure AI Foundry, covering: 🔍 Evaluator Categories and Use Cases RAG Evaluators (Retrieval-Augmented Generation) Retrieval: Measures relevance of retrieved documents. Groundedness / Groundedness Pro: Assesses factual alignment with context. Relevance & Completeness: Evaluates how well the response answers the query and covers all necessary information. Agent Evaluators Intent Resolution: Accuracy in understanding user intent. Task Adherence: Measures how well the agent completes assigned tasks. Tool Call Accuracy: Validates correct tool invocation and usage. General Purpose Evaluators Fluency: Natural language quality and readability. Coherence: Logical consistency and flow. QA: Comprehensive quality checks for question-answering tasks. 📊 Operational Observability Real-time metrics: latency, token usage, error rates. Visualizing evaluator outputs in the Foundry Observability Dashboard. Connecting evaluations to traces for root cause analysis. By the end of this session, attendees will understand how to design, implement, and scale evaluator-driven observability in AI Foundry to build robust, transparent, and production-grade AI systems.

Bio

Senior Azure Cloud & AI Consultant at Microsoft

Abstract

The main focus here is to truly teach the audience the mindset of attack/defence, especially in an environment as underexplored as GPUs and HPC as a whole. We will present real-life examples and have a small ecosystem in the cloud to demonstrate and raise awareness about this crucial topic. We will show here why the offensive mindset is a differentiator and how it helps us guide and build new security paths in Kubernetes, focusing on tools of our cloud-native ecosystem and creating a new path for newcomers who will eat GPUs for Breakfast and the rest of us.

Bio

Mart is an Infrastructure Security Manager , where he enjoys managing various engineers who teach him every day how to break things and become a better manager and engineer. Mart began his journey in cybersecurity trying to understand why so many people liked prime numbers. From there, understanding how these numbers ended up in the clouds and even inside processors became a fascination. He enjoys playing with obscure technologies and trying to be a chef in his spare time.

Abstract

In today's digital space, downtime adversely impacts customer trust and can lead to lost revenue. This session will show you how to rethink your infrastructure monitoring with logs, metrics, and traces for complete visibility. By utilizing better practices, applying machine learning for anomaly detection, and using AI assistance in place of human interactions with data, organizations can prevent and remediate issues before they lead to downtime, and to continue operating in reliable digital ecosystems. This session uses OpenSearch to present these platform-agnostic concepts.

Bio

I am a Specialist Solution Architect for Analytics at Amazon Web Services, specializing in Amazon OpenSearch Service. My role is to guide customers through essential concepts and design principles for optimal cloud-based analytics, search and GenAI deployments. When not architecting solutions, I enjoy discovering new destinations and exploring diverse culinary experiences.

Abstract

In complex platform environments, incidents don’t always begin with loud failures, they start with subtle drifts, degraded assumptions, and silent breaks in mental models. This talk tells the story of a narrative-driven coordination system inspired by Resilience Engineering principles, and how we can build a more adaptive, human-aware internal platform. We’ll explore building distributed container systems using k0rdent and implementing a resilient infrastructure tackling surfaced weak signals, mapped operator intent, and supported resilient behaviors not just at runtime, but across platform workflows. Along the way, we’ll unpack real patterns from platform SRE practice: designing for graceful failure, enabling improvisation, and embedding recovery capacity into developer tools. Whether you’re building golden paths or operating shared infrastructure, this talk will show how narrative, adaptation, and feedback loops are key to resilient platforms and the teams who build them.

Bio

Prithvi Raj is a Community Manager & Developer Advocate at Mirantis working with the Open Source Program Office team. He is a CNCF Ambassador with 5+ years of experience who helped scale a CNCF incubating project in LitmusChaos and runs the Platform & Resilience Engineering Meetup group. He is an active member at SIG Project Reviews and is currently leading the community efforts for the k0s project, k0rdent, OpenSDN, and the other OSS initiatives Mirantis is contributing to.

He is a co-organizer to KCD Bengaluru, and previously did Chaos Carnival, LitmusChaosCon and KCD Chennai. He was previously at Harness, ChaosNative, and MayaData.

Abstract

If you've worked with Terraform in production, you've likely encountered the pain of managing multiple environments, duplicate configuration files, and complex remote state setups. Terragrunt solves these common Infrastructure as Code challenges by providing a thin wrapper around Terraform that promotes DRY (Don't Repeat Yourself) principles and simplifies configuration management. This talk will take you from Terragrunt zero to hero, covering foundation and setup topics including what Terragrunt is and why it exists, key differences between Terraform and Terragrunt, and installation and initial configuration. Prerequisites include basic AWS knowledge and basic Terraform knowledge.

Ceyda is a Cloud and Platform Engineer at Sufle who transitioned to DevOps after working as a backend developer with Django and microservices. She completed her Master's in Software Engineering and holds AWS certifications including Solutions Architect Associate and HashiCorp Certified Terraform Associate. She has extensive hands-on experience using Terraform and Terragrunt for large production workloads across AWS environments, and actively contributes to the tech community through mentoring and public speaking.

Bio

Ceyda serves as a Cloud and Platform Engineer at Sufle. She has previously worked as a Software Developer at different companies. She has experience particularly in technologies such as AWS, Kubernetes, and Python. She completed her Master's degree in Software Engineering at Boğaziçi University and graduated from Sabancı University with a Bachelor's degree in Computer Science and Engineering. She holds a HashiCorp Certified Terraform Associate certification and has been using Terraform and Terragrunt for big production workloads.

Abstract

SREs are the guardians of reliability — we build for failover, redundancy, and scale. But in today’s AI-native systems, there’s a quiet shift happening: automation scripts, AI-driven agents, and machine identities are running critical operations with increasing autonomy.

They auto-remediate incidents, make scaling decisions, deploy configurations, and even patch systems — often with elevated permissions and little human oversight.

But here’s the problem:

• What happens when these agents act on bad data?
• Or when a misconfiguration triggers a cascade of failures?
• Who holds accountability when an over-permissioned automation silently breaks production?

In this session, I will expose the blind spot most SRE teams don’t realize they have — the growing influence of non-human actors in reliability engineering. I will share real-world examples of automation gone wrong, highlight how over-trusted machine identities quietly amplify risk, and explain why SREs need a governance mindset toward automation.

I will discuss:

• The evolving role of AI agents and automation in SRE workflows
• How machine identities can quietly become single points of failure
• Why over-permissioned bots and lack of observability threaten reliability
• Practical steps for introducing identity governance, privilege boundaries, and automation observability into SRE practices
• How to chaos-engineer for AI agents and validate fail-safes

This is not a security talk — it’s a reality check for SREs operating in environments where automation does more than assist; it decides. If you are an SRE, platform engineer, or operations lead, this talk will challenge you to rethink your approach before your AI helpers become your biggest liability.

Bio

Cynthia Akiotu is a Cybersecurity Architect and Identity Specialist with deep expertise in Identity & Access Management (IAM), data governance, and Zero Trust implementation. She focuses on securing environments where AI agents, autonomous models, and machine identities are rapidly expanding the attack surface. With experience across diverse sectors, Cynthia has led projects on secure cloud adoption, privileged access management, insider risk mitigation, and identity governance for AI-driven environments. She holds several industry certifications, including Microsoft Certified: Identity & Access Administrator and Microsoft Certified: Information Security Administrator. Her thought leadership extends to academic and industry publications, contributing to works on AI-enabled regulatory compliance and digital skills development in security. Beyond her professional work, she volunteers as Cyberhero Coach Cynthia, promoting cybersecurity awareness and safe digital habits among children and communities. Driven by a commitment to secure innovation, she champions identity governance as a foundation for trust in AI-native ecosystems.

Abstract

In high-scale environments, metrics cardinality isn’t just a resource concern, it’s an architectural challenge. Left unchecked, it can impact performance, query latency, and even system stability. This talk takes a deep technical dive into how VictoriaMetrics enables advanced observability practices with a strong focus on cardinality management. I’ll explore how to design efficient and scalable scrape configurations using Prometheus-compatible jobs and exporters, optimize your label strategies, and use built-in cardinality analysis tools within VictoriaMetrics to identify and mitigate high-cardinality patterns early. The session also covers integration with Grafana open source for visualizing metrics in a way that supports signal clarity and operational response, as well as setting up practical alerting strategies that minimize noise while ensuring fast issue detection. By the end of the talk, the audience will understand the engineering trade-offs of high-cardinality metrics and how to detect them, how VictoriaMetrics handles storage and querying at scale, how to build resilient and low-overhead scrape configs with Prometheus compatibility and how to use Grafana open source to highlight cardinality hot spots and improve alert signal quality.

Bio

Diana is a DX Engineer at VictoriaMetrics. She is passionate about Observability, machine learning. She is an active contributor to the OpenTelemetry CNCF open source project and supports women in tech.

Abstract

In the ever evolving AI landscape, organizations are faced with the choice between open-source versus closed-source models. While many developers find it easier to get started in the closed source ecosystem, they quickly realize it is ultimately more expensive, inefficient, and doesn’t have the security and controls required for their applications. Alternatively, open source models have quickly caught up to their closed counterparts and now deliver a cheaper and just as accurate solution that provides organizations with the security they need to run their operations. In addition, the pace of the open source community is iterating at speeds that are beyond belief and not only improving the accuracy of these models, but making them run even faster, especially on hardware that is better suited for AI beyond the GPU. With new open source models dropping every week and an active community fine-tuning even better versions daily, open source is making AI a commodity, with a wide range of cloud services for developers to choose from that make it even easier for them to use. Join Amit Kushwaha, Director of AI Engineering, SambaNova, as he breaks down for attendees the advantages of open-source models, why they are critical for fast product iteration, and how organizations can use them to tap into the best minds globally, accelerate their pace of innovation, and position themselves to shape industry standards while continuously advancing enterprise-grade solutions.

Bio

Diana is a DX Engineer at VictoriaMetrics. She is passionate about Observability, machine learning. She is an active contributor to the OpenTelemetry CNCF open source project and supports women in tech.

Abstract

A real-world case study of a B2B fintech startup, powered by a lean team and a fully serverless, AWS-native architecture. Continuous Deployment was implemented from day one, enabling dozens of automated releases per week with minimal overhead. This delivery setup was tightly integrated with Continuous Discovery practices - customer interviews, lightweight experiments, and outcome-driven KPIs - to ensure engineering velocity translated into real product value. Together, these practices created a compounding feedback loop that accelerated product–market fit, improved competitive positioning, and drove measurable financial results. The session will cover the tools, processes, and cultural choices that made this possible, in a journey that ultimately led to a successful private equity exit in under four years.

Bio

I'm the Co-Founder and CTO of countX, a SaaS platform that enables SMEs to navigate cross-border e-commerce with ease by automating VAT compliance, accounting, and reporting. With over a decade of experience in software engineering and product leadership, I specialize in building scalable platforms, growing high-performing teams, and delivering customer-centric solutions.

Previously, I founded Sfeeda, an innovative platform for real-world developer assessments, and held engineering leadership roles at AUTO1 Group, where I led multiple teams through large-scale technical transformations. My approach combines lean product development, continuous discovery, and continuous delivery to create empowered product teams that move fast and deliver value.

Abstract

We’re in the midst of an AI revolution—and APIs are its unsung heroes. While LLMs and AI agents grab headlines, it's APIs that power their ability. Behind every AI-generated insight, recommendation, or automated task is an API call connecting the model to the tools, services, and data it needs to get the job done. As AI systems evolve from passive assistants to autonomous agents capable of decision-making and execution, APIs have become the essential infrastructure enabling this transformation. They are no longer just integration tools—they are the action layer of AI. In this talk, we’ll explore how APIs are shaping the future of intelligent automation. Using real-world examples from across industries, we’ll examine how companies are leveraging APIs to orchestrate multi-step workflows, access real-time data, and drive operational efficiency with AI. Organizations with robust, scalable, and discoverable API ecosystems will not only keep up—they’ll lead. If AI is the recipe, APIs are the ingredients. It's time we start treating them that way. What You’ll Learn: - The shift from human-first to machine-first consumption patterns in API design - Emerging standards that are streamlining AI-API interactions - Strategies to future-proof your API ecosystem for the intelligent systems of tomorrow

Bio

Pooja Mistry is a passionate Senior Developer Advocate at Postman, where she champions the power of APIs by creating technical content, leading workshops, and building developer communities around the world. With a strong background in cloud, automation, and open-source technologies, she brings over a decade of hands-on experience in software engineering and advocacy. Before joining Postman, Pooja held several impactful roles at IBM, including leading developer outreach for IBM Cloud and contributing to healthcare innovation at IBM Watson Health. She's also been an active community leader, having organized Startup Weekend events to empower entrepreneurs and technologists across Boston.

Abstract

Observability is the ability to measure the current state of a system. SREs are familiar with OpenTelemetry signals and how they can be used to diagnose issues. Yet in the frontend world, we're behind the curve.

Join me as I dive into the current state of OpenTelemetry for Web. I'll cover a simple example of logging, metrics and tracing using OTel web, specifically the Client Instrumentation libraries, and the types of metrics and traces we should collect to diagnose performance issues and client usage in UIs.

Bio

Carly is a Developer Advocate and Manager at Elastic, based in London, UK. Before joining Elastic in 2022, she spent over 10 years working as a technologist at a large investment bank, specialising in Frontend Web development and agility. She is an agile evangelist, UI enthusiast, and regular blogger.

She enjoys cooking, photography, drinking tea, and chasing after her young son in her spare time.

Abstract

Two Buzzwords Finally Meet! How should we manage AI in large organizations? What are the "services" developers need to add AI to enterprise apps, and what role do platform engineers take? Where do data scientists fit in? How does MCP, A2A, and whatever the latest AI API is fit in? There are so many questions because the field of managing AI in enterprises barely exists. We need to figure it out quickly however, before we see a repeat of a past patterns, like shadow AI and unmanageable apps in production. Based on real-world examples, this talk will go over what we currently know about using platform engineering to help developers use AI.

Bio

Michael Cote studies how large organizations get better at building software to run better and grow their business. His books Changing Mindsets, Monolithic Transformation, and The Business Bottleneck_ cover these topics. He's been an industry analyst at RedMonk and 451 Research, done corporate strategy and M&A, and was a programmer. He also co-hosts several podcasts, including Software Defined Talk. His daily-ish newsletter is at newsletter.cote.io.

Abstract

With our traditional operations setup supporting monolithic systems – as the tech implementation scaled, there was a proportionate increase in costs to support and maintain the systems. This had been primarily due to segregation in organization setup to deliver business outcomes. While one part of the tech organization was focused on build, the other part was focused on maintain. With differing goal-post and motivation to deliver around collective business impact, the focus was divergent which led to undesirable customer experience and unsustainable costs. As the tech landscape became more distributed and decoupled, we saw this as an opportunity to introduce SRE adoption, where we managed to build an integrated setup with product teams and embedded SRE taking full accountability from build to maintain for the product even as the implementation scaled across multiple markets. While this setup has proved to be impactful over the last couple of months, it has been a continuous effort to tie Service Level Indicators and Objectives (SLI/SLO) to Business KPI's thereby showcasing direct impact to business. Will delve into ways organizations can successfully adopt SRE thereby building an integrated and impactful setup.

Bio

Mitul Jain Digital Transformation Leader | SRE & DevOps Expert | Global Technology Executive With over 23 years of global leadership experience, a recognized expert in Site Reliability Engineering (SRE), DevOps, and enterprise-scale automation. I have led transformative initiatives across industries, reshaping traditional IT operations into modern, business-aligned engineering practices. I have built and scaled global SRE organizations, integrated reliability engineering into digital value streams, and enabled Build-to-Operate models that drive measurable business outcomes—including zero P1 incidents during peak periods and significant cost optimizations. As a passionate advocate for innovation and community building, I founded SRE Community of Practice within the organization and has led strategic alliances with platform providers to enhance service capabilities and go-to-market strategies. I often speak on topics including digital transformation, reliability engineering, DevOps at scale, and the future of IT operations. His insights are shaped by hands-on experience across Consumer Goods, BFSI, Telecom, Manufacturing, and Travel & Hospitality.

Abstract

Multi-cloud adoption breaks the one-size-fits-all promise of traditional PaaS. This talk dives into how we built an internal developer platform that balances flexibility, control, and speed across clouds.

Bio

I’m Ashish Upadhyay, a platform engineering leader passionate about building internal developer platforms that seamlessly bridge AWS, Azure, and on-prem systems. My focus is on simplifying complexity, enabling speed, and giving developers the tools they need to thrive. Outside of work, I am fascinated by astronomy.

Abstract

As systems grow more complex, engineering teams must move beyond guesswork when resolving production issues and instead focus on delivering consistent, measurable user value. This talk explores how OpenTelemetry empowers teams to define, measure and improve Service Level Objectives with precision and confidence.

Drawing on real-world experiences of a software team navigating the challenges of legacy code, cloud migration and production instability, this session demonstrates how OpenTelemetry helps teams gain the insights they need to reduce Mean Time To Resolution, understand root causes and proactively improve system reliability.

We’ll dive into hands-on examples using the OpenTelemetry Demo Application and show how to help turn technical metrics into business-aligned SLOs that actually mean something. Whether you're an SRE, developer, or engineering leader, this talk will give you the clarity—and tools—to move from reactive firefighting to proactive service reliability.

Bio

Martin McLarnon is an engineer with almost 30 years experience working in many different roles within IT. He has worked as a Network Manager, Software engineer, Cloud Architect, Engineering lead and worked as an engineering consultant before joining Coralogix.

Abstract

One of the common practices for improving the query speed in Prometheus is to create recording rules for commonly used queries. While this usually works great, recording rules have a cost: The raw metrics still need to be stored in the Prometheus, even if we don't need them Recording rule need to be executed on interval basis, putting extra pressure on the Prometheus Adding a recording rule always means storing&processing more data

But what if we could do the pre-aggregation before metrics get into the Prometheus? Can we aggregate on scrape time? Or could clients do the aggregation before pushing data to Prometheus? At VictoriaMetrics we already started working on something we call Stream Aggregation. An ability for metrics collector to perform in-memory aggregations before forwarding data elsewhere (Prometheus including). It has many challenges related to time series nature of aggregated samples, network delays, horizontal scaling, etc. I'd like to share our experience with the community, as this may be a life-quality improving feature.

Bio

Roman is a software engineer with experience in distributed systems, databases, monitoring, and high-performance microservices. Roman's passion is open source and he's proud to have contributions to Prometheus, Grafana, and ClickHouse. Currently, Roman is working on the open source time series database and monitoring solution VictoriaMetrics.

Abstract

Chaos Engineering is often misunderstood as simply “breaking things on purpose.” This talk challenges that perception and repositions Chaos Engineering as a critical pillar of reliability and resilience engineering. Rather than focusing on failure injection alone, we explore how to leverage existing knowledge, validate known truths, and foster confidence in complex systems.

In the first part, we deconstruct common myths around Chaos Engineering and reframe its core principles. Learn how aligning chaos practices with reliability goals can transform the way your organization perceives and applies these techniques—by emphasizing structured validation over blind experimentation.

The second part brings theory into practice with a hands-on framework that reimagines chaos experiments as self-feeding, iterative loops—mirroring the scientific method. We introduce the concept of continuous verification, drawing parallels with integration testing, and show how Chaos Engineering can seamlessly integrate into the Software Development Lifecycle (SDLC) through a shift-left approach.

The session wraps up with a visual framework for implementing a sustainable Chaos Engineering strategy, including how to evolve gamedays into repeatable, hypothesis-driven validations that scale with system changes.

Whether you're just exploring Chaos Engineering or looking to mature your reliability strategy, this talk will leave you with actionable insights, a modernized mindset, and a clear path to operational resilience.

Bio

I am a software engineer with an attrition for low-level and Linux internals who joined Datadog 7 years ago in the SRE group. After developing the now open-source Chaos Controller project meant to provide large-scale fault injection to engineering teams, I built the Security Chaos Engineering group now made of 3 teams and 20 people for which I act as the Engineering Manager. I am also part of the Core Incident Commanders group involved in all high severity outages to help with internal coordination and external communication.

Abstract

Sustainability is increasingly becoming a priority in the Information Technology sector, which has fueled the demand for energy-efficient solutions in all computing environments. Effective management of resources in Kubernetes environments requires proper monitoring and optimisation of power usage. Kepler (Kubernetes-based Efficient Power Level Exporter) solves this problem by offering a solid solution for energy monitoring at the pod level. Using software counters, tailored machine learning models, and the Cloud Native benchmark suite, Kepler provides accurate energy consumption estimations and detailed reports on power usage. Developers and system administrators can thus make informed choices towards environmentally friendly and energy-efficient Kubernetes operations.

Bio

Mayank Goyal is a Senior Site Reliability Engineer at Okta, specialising in maintaining Workflows systems' scalability and reliability. Before working at Okta, he honed his DevOps & Software engineering skills at Zoom and RedHat, where he grew deeply passionate about automation. While at RedHat, he was instrumental in automating and configuring the Release Pipeline in RHELWF. Apart from his professional obligations, Mayank is a passionate supporter of open-source technologies and likes being on the cutting edge. He loves connecting with the technology community, exchanging thoughts, and learning from others.

Abstract

AMBA is not my new AI assistant, but it does help me know what's going on in Azure. Rapid cloud adoption has created a wealth and yet a gap on the essential visibility within environments and platforms. In this talk we will cover how Azure Monitor Baseline Alerts (AMBA) can help you to kickstart your proactive and informed approach to cloud with Microsoft.

Bio

Jamie has been working in IT for more than 10 years in a variety of roles from 1st line support, Managed services and most recently within the DevOps space.

This has helped to create a broad range of knowledge that he is keen to share with others. This passion has led him to co-found a user group community with the aim to help others learn new skills and share their knowledge.

Abstract

In our world where everything is code, reliability extends beyond clean and reliable code running on the right infrastructure. It requires a robust sociotechnical system, the dynamic interplay between social and technical components. Our North Star is an engineering culture, built on shared beliefs, practices and behaviours that shape how we operate, solve problems, collaborate, innovate and continuously learn. But how do we achieve this engineering culture, where innovation and success are driven by collaboration, trust, autonomy, and passion? What is our code of conduct and how does it propel us forward? In this talk, we will take you through our 5-year journey in Reliability Advocacy. We started out as a group of 10 enthusiastic engineers from different platform and enablement teams who wanted to share knowledge of our reliability product offering and practices across the organization. We are now a household name, run our annual Reliability Event conference which are attended by 300+ engineers every year and our reliability trainings are part of the curriculum for all new joiners and we've guided hundreds of engineers on their path to defining SLIs and SLOs. In part, thanks to our reliability advocates, reliability is now one of our engineering pillars for years to come shaping our organization in the process, raising our general availability rating from 99.54% to 99,87% in the process. We will share what worked for us, what didn’t, and how we gradually embedded ourselves into our large, regulated, and risk-averse organization with over 15.000 engineers. We will reveal the code that helped us shape our SRE practices and make this transformation possible. In doing so, we will share a 5-step plan to start a reliability advocacy function within your organization.

Bio

Stephan Mousset is the Global Product Manager for the Performance & Resilience Engineering Platform at ING, where he also serves as Lead Reliability Advocate. With nearly two decades of experience at ING, Stephan has led major efforts in performance testing automation, SLI/SLO adoption, and reliability engineering at scale. He is an active voice in the DevOps and SRE communities — co-organizing DevOpsDays Amsterdam, Site Reliability Engineering NL, and ING’s internal Reliability Event. Stephan has shared his insights at multiple conferences, including: - SLOconf 2022: ING’s global SLO rollout and lessons learned - SRE NL Meetup at ING (2023): Building a platform for SLO-driven performance engineering - DevOpsDays Berlin 2024 (Ignite): "Unleashing DevOps Magic in Performance Testing & Analysis Automation" His talks focus on practical, platform-enabled approaches to making reliability and resilience part of everyday engineering — through automation, observability, and culture.

Abstract

In the age of AI, efficiency and budget-cuts, the idea of hosting internships or early-career grads may seem in redundant - but this rhetoric couldn't be more misplaced! This session aims to demystify working with students, graduates and educational institutions. Instead, come prepared to discover the range of benefits businesses gain from collaborating in this way, in addition to resources and funding opportunities. Ever considered taking on an intern, but unsure where to start? Do you have projects that would thrive with the keen support of interns? … Then this talk is for you!

Bio

Alecia Cotterell is Head of Careers at Teesside University London, which specialises in Business, Computing and Games & Animation. She developed the campus-wide Employability Curriculum, setting all students up for graduate success. Alecia also leads the Internship Module and Digital Skills development. In addition, she manages Personal Tutoring and is the campus' Disability Co-ordinator, supporting inclusivity for all students.

Abstract

When Spotify scaled from millions to hundreds of millions of users, we discovered that our carefully crafted abstractions—designed to simplify our systems—had become our biggest operational liability. New engineers could ship features but couldn't debug failures. Our beautiful, clean interfaces masked the very complexity that seasoned engineers needed to understand during critical incidents. This talk chronicles how internal fragmentation during hyper-growth exposed a fundamental paradox in software engineering: the tension between helpful abstraction and dangerous oversimplification. Through real examples from Spotify's infrastructure evolution, you'll discover why traditional approaches to hiding complexity often backfire at scale.

What You'll Learn - The Abstraction Trap How well-intentioned design decisions can create cognitive bottlenecks during incidents - Graduated Disclosure A practical framework for building abstractions that reveal complexity progressively - The Teaching Principle Why the best abstractions educate rather than obscure - Scaling Operational Knowledge Concrete strategies for maintaining system understanding as teams grow

Who Should Attend This session is valuable for: - Software engineers - Platform teams - Site Reliability Engineers (SREs) - Engineering leaders Especially those who struggle with: - Onboarding engineers to complex systems - Maintaining operational excellence during rapid growth - Balancing developer productivity with system transparency - Building abstractions that enhance rather than hinder incident response

Key Takeaways You'll leave with actionable insights for designing systems and tooling that scale both technically and cognitively—ensuring your abstractions become force multipliers rather than barriers to understanding.

Bio

Stuart is a sought-after speaker, TEDx presenter, and champion for sustainable software. He's passionate about the impact of software and AI on the climate and empowers developers to build a more sustainable future. As a leading expert in programmability and DevOps, he frequently graces industry stages worldwide, sharing his knowledge and inspiring others. He lives in Lincoln, England, with his wife, Natalie, and their son, Maddox. He plays guitar and rocks an impressive two-foot beard while drinking coffee.

Abstract

The promise of cloud computing—elasticity, agility, and reduced operational overhead—often hinges on the concept of stateless application design. Yet, every meaningful application relies on persistent data, creating a fundamental tension: how do you manage and serve "state" effectively in a "stateless" world? This 30-minute session will equip architects, developers, and operations professionals with the knowledge to navigate this critical challenge.

We'll begin by dissecting the inherent paradox of state in the cloud, understanding why traditional database architectures often falter in highly distributed, ephemeral environments. The talk will then pivot to practical strategies for architecting data serving layers, exploring a spectrum of database types—from traditional relational databases like PostgreSQL and MySQL to modern NoSQL solutions such as Aerospike, CockroachDB, and DynamoDB, and even specialized options like data warehouses and graph databases. We'll discuss the unique scaling characteristics and ideal use cases for each.

A significant portion of the talk will be dedicated to the critical decision between leveraging fully managed Database-as-a-Service (DBaaS) offerings (like AWS RDS, Azure Cosmos DB, Google Cloud Spanner) versus deploying and managing databases yourself on cloud infrastructure. We'll weigh the pros and cons of each approach, considering factors like operational burden, customization needs, cost implications, and vendor lock-in.

Finally, we'll dive into the core pillars of successful cloud data serving:

Scaling: Understanding horizontal vs. vertical scaling, sharding strategies, and auto-scaling mechanisms.

Reliability: Implementing high availability (HA) architectures, disaster recovery (DR) plans, robust backup and restore procedures, and data consistency models.

Day-to-Day Management: Best practices for monitoring, patching, upgrades, performance tuning, and cost optimization in a cloud context.

Choosing the Right Type: A framework for evaluating your application's specific requirements (data model, access patterns, consistency needs, transactionality, scale) to select the most appropriate database technology and deployment model.

Attendees will leave with a clearer understanding of how to design, deploy, and manage data serving solutions that are not just cloud-compatible, but truly cloud-native, ensuring their applications remain performant, reliable, and scalable in the ever-evolving cloud landscape.

Bio

Harel Safra is a Data Platform Engineering Team Lead at Riskified, a Fintech company focusing on fraud prevention. His team manages the online data platform supporting real time fraud detection systems. He has worked in various Data And Infrastructure roles throughout his career, always striving for simpler ways to achieve scalable, optimally performing systems

Abstract

Engineers have long believed that managed Kubernetes services in the cloud would liberate them from vendor lock-in, but the reality is far from it. Most container platforms fall short of being feature-complete, leaving a trail of external concerns that can derail even the smoothest migration.

Join us on a journey as we take a demo application from AWS to GCP and expose the hidden pitfalls that await Kubernetes app portability. We'll distill essential software development and system architecture best practices to help you future-proof your Kube applications, ensuring they're free to roam across cloud providers without getting stuck in vendor-specific quicksand.

Additionally, we'll reveal configuration and components within the Kubernetes ecosystem that can be leveraged to effortlessly bypass platform lock-in constraints. Don't get caught off guard – learn how to safeguard your app's portability today!

Bio

Federico Fregosi is a Staff Software Engineer (SRE) at Wise with deep experience in cloud, data, and machine learning platforms. Former VP of Engineering at Contino and Principal Consultant at OpsGuru, he has led large-scale migrations, built resilient systems, and shaped engineering cultures in high-growth organizations.

Abstract

Modern observability is built on a flawed foundation: three siloed pillars - logs, metrics, and traces - each powered by different engines with separate query models, storage formats, and operational costs. Users are forced to manually correlate across systems, accept duplication, or pay high SaaS bills. But what if observability is just a data problem? One that needs a general-purpose solution instead of purpose-built compromises? This talk argues that true observability requires fast, high-cardinality queries over unsampled data at scale and at low cost. Traditional search and metrics engines were not designed for this, but column-oriented databases are. We introduce ClickStack, a fully open source, OpenTelemetry-native observability stack built around ClickHouse. It provides fast, flexible querying and efficient storage, enabling real-time visibility without compromise.

Bio

Dale McDiarmid is a Senior Principal Technologist at ClickHouse with over a decade of experience in data infrastructure, performance engineering, and product strategy. Based in Lisbon, he has held key roles at Grafana Labs, Elastic, and Attivio, specializing in database optimization, observability, and technical marketing.

Abstract

This talk showcases seamless integrations between ClickHouse and key tools like OpenTelemetry, Prometheus, and Grafana, fostering greater interoperability. It drives innovation in observability by addressing the challenge of fragmented monitoring systems through a unified telemetry solution.

Bio

Joshua is a seasoned software developer with over a decade of experience, specializing in a broad range of topics including operations, observability, agile methodologies, and accessibility. His passion for technology is matched by his enthusiasm for sharing knowledge through public speaking. Currently, Joshua serves as a Developer Advocate for Altinity, where he creates educational content on ClickHouse and OpenTelemetry. Additionally, he is an active contributor to the OpenTelemetry project, helping to advance the field of observability in software development.

Abstract

Have you ever wondered what observability really means and how it can be useful, even if you’re just starting out? In this session, we’ll explore the basics of observability through a fun, relatable project: monitoring the health of plants using simple IoT sensors and Grafana dashboards. Using a hands-on example, I’ll share how I teamed up with my 8-year-old daughter to build a soil moisture sensor project. We wired a sensor to an ESP32 board, collected moisture data, sent it to Prometheus, and visualized it using Grafana, a powerful but beginner-friendly tool for creating dashboards. Along the way, I’ll explain key observability concepts like metrics, monitoring, and dashboards, and break down technical jargon with gardening analogies that make it easier to grasp. This session requires no prior experience, just curiosity and a love of learning. By the end, you’ll see how observing your systems (or your plants!) through data can be both empowering and fun, and how Grafana makes it easy to bring data to life visually.

Bio

Marie Cruz is a Software Tester with over 10 years of experience and currently works as a Senior Developer Advocate for Grafana Labs. She is also the co-author of Contract Testing in Action, the first book dedicated solely to contract testing, and an international speaker. She has previously worked as an engineering manager responsible for driving continuous testing and quality improvements, and as a principal engineer, she focused on introducing recommended practices for testing and test automation frameworks.

Abstract

Let's take a look at some of the basic NGINX metrics to monitor and what they indicate. We start with the application layer and move down through process, server, hosting provider, external services, and user activity. With these metrics, you get coverage for active and incipient problems with NGINX

Bio

Currently providing technical evangelism for NGINX, Dave works with DevOps, developers and architects to understand the advantages of modern microservice architectures and orchestration to solve large-scale distributed systems challenges, especially with open source and its innovation. Dave has been a champion for open systems and open source from the early days of Linux, through open distributed file systems like XFS, GFS, and GlusterFS to today's world of clouds and containers. He often speaks on topics such as the real-world issues associated with emerging software architectures and practices, on open source software and on creating new technology companies. Dave has spoken on technical topics like distributed request tracing, modern monitoring practices, open sources projects from both corporations and foundation views, and on how open source innovations powers todays world. Dave was named as one of the top ten pioneers in open source by Computer Business Review, having cut his teeth on Linux and compilers before the phrase "open source" was coined. Well versed in trivia, he won a Golden Penguin in 2002. When he's not talking, you can find him hiking with his trusty camera, trying to keep up with his wife.

Abstract

Let's be honest: writing correct concurrent and distributed code is hard. Race conditions? Deadlocks? They're the bane of modern software development, especially with tools like Kafka, Kubernetes, and microservices. Testing helps, but it rarely finds all the subtle, timing-dependent bugs. Imagine having a way to mathematically verify your design's logic before you code it. That's TLA+. Developed by Turing Award winner Leslie Lamport, it's a language for specifying and checking algorithms, especially concurrent ones. The catch? Most intros are dense and theoretical. This session is different. It's TLA+ for the rest of us – the software engineers in the trenches. We'll skip the deep math and focus on practical application. You'll learn what TLA+ is, how to model simple concurrent interactions, and how it can save you from deployment disasters by catching design flaws early. Walk away ready to explore how formal methods can become your secret weapon for building bulletproof systems. Key Takeaways: - Recognize the inherent difficulties in validating concurrent designs. - Get a practical, hands-on feel for TLA+ concepts and syntax. - Understand how TLA+ helps prevent race conditions and deadlocks at the design stage. - Be inspired to apply formal verification techniques to real-world problems.

Bio

Consulting software architect / lead dev. Specialized in backend performance optimization. Wrote THE French book on Go: https://osinet.fr/go/en/ - Main non-tech business interests: media and e-commerce. Also, playing bass guitar with lots of distortion when I get the chance.

Abstract

Your AI might be the biggest insider threat you’ve ever deployed without even knowing it. As large language models (LLMs) become embedded in cloud-native apps and infrastructure, a new kind of risk is emerging: data leakage through inference. With the right prompt, an attacker can extract sensitive data, proprietary logic, or even credentials from your model’s responses bypassing traditional cloud and application security. In this talk, we’ll explore real-world examples of prompt injection, model inversion, and inference-time exfiltration attacks. You’ll learn why AI models running in Kubernetes, serverless, or SaaS environments introduce hidden exposure, and what to do about it. Whether you’re a security engineer, cloud architect, DevSecOps lead, or AI/ML practitioner, this session will equip you to recognize and defend against one of the fastest-evolving threats in modern cloud systems. Because in the age of AI, the model doesn’t need to be hacked to become the leak.

Bio

Meletius Igbokwe is a cybersecurity-focused Modern Workplace Engineer with a strong passion for protecting organizational data and identities in today’s cloud-driven world. With expertise spanning cloud security, infrastructure, and automation, he is dedicated to helping organizations adopt modern technologies while maintaining a strong security posture. A professional member of BCS, The Chartered Institute for IT, and a holder of a Master’s degree in Cloud and Network Security from the University of Bolton, Meletius is deeply invested in advancing cybersecurity through research, innovation, and real-world application. His interests include the intersection of AI and cybersecurity, digital forensics, identity security, and securing enterprise environments using industry best practices. Meletius has contributed to several cloud transformation initiatives, using DevOps methodologies to enhance secure deployment and operational efficiency. He prioritizes resilience, compliance, and scalability in every solution he delivers.

Abstract

Site Reliability Engineering was never meant to be about firefighting, yet too many teams find themselves stuck in an endless cycle of pages, postmortems, and quick fixes. Why? Because SRE is full of hidden traps — patterns that look like best practices on the surface but slowly erode reliability, burn out engineers, and stall progress.

In this talk, we’ll expose the 7 Deadly Traps of SRE, from the obsession with chasing “five nines,” to the cult of on-call heroism, to the false comfort of tooling and checklists. For each trap, we’ll unpack why it’s so seductive, how it quietly sabotages your team, and what to do instead.

You’ll walk away with a clearer lens on the pitfalls holding SRE organizations back, and a practical playbook to help your team escape firefighting mode and reclaim the true purpose of SRE: building systems - and cultures - that are resilient, scalable, and human-friendly.

Bio

Miko Pawlikowski is an SRE Author and a platform engineer at Quadrature. He has led large-scale infrastructure and SRE initiatives at Citadel and Bloomberg, with deep expertise in Kubernetes, cloud computing, and chaos engineering. Passionate about building resilient systems and communities, he brings together engineers worldwide through conferences and media projects.

Abstract

Today’s observability platforms are often vertically integrated—binding data storage, query, and visualization layers into a single stack. This tight coupling drives up costs, makes integrations painful, and slows teams down. But it doesn’t have to be this way. In this talk, we’ll explore how SRE teams can benefit from a more modular approach to observability—one inspired by the evolution of Business Intelligence. Just as BI stacks evolved to separate ETL, data warehouses, and dashboards, observability stacks can be designed around clear boundaries: interoperable tools, technology-neutral query layers, and plug-and-play storage. You’ll learn why decoupled observability architecture is essential for cost control, agility, and tool flexibility—and how to move toward a stack that meets the real-world needs of today’s SRE teams.

Bio

Peter Marshall is a technology leader and community builder with a background in developer relations, data architecture, and digital transformation. As Director of Developer Relations at Imply, he leads programs that grow and engage the global Apache Druid community through education, support, and events. With experience across startups, enterprises, and the public sector, Peter brings a blend of technical expertise and strategic vision to help organizations connect with developers and drive impact through open source.

Abstract

While LLMs make AI agent creation seem simple, building reliable production solutions requires structured methodology. This talk shares key insights from my year-long project developing an AI-powered root cause analysis system for Kubernetes clusters. I'll present three strategies that transformed our approach from experimentation to predictable success: Automated Iterative Context Building: Building context progressively rather than overwhelming LLMs with data upfront Standardized Prompt Architecture: Creating structured templates that simplify implementation across use cases Optimized Data Presentation: Formatting data to significantly improve accuracy and reduce hallucinations I'll demonstrate these through our "Kubernetes Deployment Failure Investigation" system, where our LLM approach outperformed traditional coding by considering a wider range of factors. Attendees will gain practical frameworks for systematic LLM application development applicable within and beyond Kubernetes, with opportunities to share experiences in this rapidly evolving field.

Bio

Andrey is the Open Source Dev Leader at Komodor, a startup building the next-gen troubleshooting platform for Kubernetes. With over 20 years of experience in the dev space, Andrey is known as the founder of JMeter-Plugins.org, an open source project which he led to its undisputed status as the premier source of JMeter plugins in the world. He’s also an avid speaker who loves talking about Open Source (could you have guessed?!), performance testing, all things automation, and his favorite programming languages Python, Golang and Java. Prior to Komodor, Andrey was the Co-founder of Up9 and before that the Open Source Evangelist at BlazeMeter.com.

Abstract

Platform engineers work hard to build great tooling and automations for developers, but often struggle to get feature teams to adopt the platform to its full potential. Meanwhile, SREs are buried in incident firefighting and can’t keep up with onboarding new services or proactive reliability initiatives. Turns out these two challenges could be solved by tackling them together. In this talk, we’ll share how we combined platform engineering and SRE into one hybrid responsibility that doesn’t just ship tooling, it helps teams actually adopt it. We’ll show how our Platform SREs make new services “reliable by default” with out-of-the-box observability, alerts, and SLOs. But for those older, messier services? We send someone in. Embedded SRE style, but for a limited time and scope. We’ll walk through how we structure these short-term embed missions, what’s worked (and what’s flopped), and how this helped adoption go way up without burning anyone out. If you’re tired of begging teams to migrate or your SREs are on the verge, this one’s for you.

Bio

Jorge is a Reliability Advocate at Rootly and the author of the Linux Foundation Introduction to Backstage (LFS142) course. He has a background in software engineering (ex-PayPal) and digital communication (UCLA). He's also a certified sommelier (CETT Barcelona).

Abstract

In this talk, we’ll explore ten frequent mistakes developers and operators make when working with Kubernetes. From misconfigured resources and insecure deployments to overlooked observability and poor scaling practices — this session will highlight real-world pitfalls and offer practical advice to help you build more reliable, secure, and efficient Kubernetes environments.

Bio

Kim is Regional Field CTO and Solution Architect at CAST AI, focused on guiding Kubernetes users to cost efficient and automated platforms. Before CAST, he was working as Cloud Architect and Consultant for Cloudical, noris network and Deutsche Telekom. He’s passionate about cloud infrastructure, automation and open source technologies. He’s an open source contributor and public speaker.

Abstract

In this session, we'll explore how to implement flexible multitenancy in Kubernetes using vCluster. You'll learn how to design platforms that can adapt to different isolation requirements, resource sharing needs, and trust models within a single cluster.

We'll explore:

• Multitenancy Spectrum
  Understanding the three dimensions of multitenancy:
• Isolation levels: physical / logical / none
• Resource sharing: dedicated / pooled / shared

• Trust boundaries: zero / partial / full

• Flexibility in Practice
  How to mix and match different multitenancy models for different teams and workloads.

• vCluster Deep Dive
  How virtual clusters enable this flexibility while maintaining strong isolation boundaries.

• Real-World Scenarios
  Live demo showing how to implement different multitenancy patterns based on actual use cases.

Key Takeaways

• Understanding the multitenancy conceptual framework and its three core dimensions
• Practical patterns for implementing flexible isolation strategies
• How to choose the right multitenancy model for different team requirements
• Hands-on experience with vCluster for creating adaptable multitenant environments

Why This Talk Matters

Most Kubernetes multitenancy discussions focus on a single approach — either namespace isolation or full cluster isolation. Real-world platforms need flexibility to accommodate different teams with varying security, resource, and trust requirements. This talk shows how to build platforms that can adapt to these diverse needs without compromising on security or operational efficiency.

Bio

An active contributor to OpenSource projects on GitHub, blogger and content creator, focusing on practical, scalable solutions in cloud-native environments. DevOps and Platform Engineering practitioner and advocate. Visit: cloudrumble.net