0.5 devopsdays-portland-2026 2026-09-08 2026-09-10 3 00:05 https://talks.devopsdays.org https://talks.devopsdays.org/media/devopsdays-portland-2026/img/logo_g2RyYvn.webp US/Pacific Room 327 Workshop 2026-09-08T13:00:00-07:00 13:00 02:00 We built a burrito-ordering AI agent, put it on Kubernetes with scoped cluster access and a live terminal, and let a room of engineers try to break it. It is tons of fun. Over three rounds they push the agent to steal customer data, deploy malicious workloads, and take the cluster down. Forkbombs anyone! The first round has nothing protecting the agent, and it usually falls apart in about one prompt. The second runs the same attacks against an ordinary CNCF platform floor, which blocks most of them, though a few run up a bill before they get stopped. In the third round attendees switch on the remaining guardrails themselves and watch how the agent's behavior and its token cost change. Running it taught us that for most shops, AI doesn't create just brand-new security problems it also makes the ones you already have more expensive. Token spend becomes its own denial-of-service vector for example. Most of what your platform already does holds up against an agent (or for an agent), and the gap that's left is smaller and stranger than people assume. This workshop covers the base guardrails as well as that enhanced gap that you must have for AI workloads. We ran this at AI Engineer World's Fair on June 29th 2026. This version covers what held up, what broke, and what we've changed since. We have made vast improvements including model triage, provisioning and sandboxing (as an option). This will be a very evolved, much more polished, and much more expanded presentation than the one we ran at AI Engineer World's Fair. This will be much more hands-on instead of just a web interface. They will actually get access to their own cluster. devopsdays-portland-2026-6809-build-a-platform-unleash-an-agent-on-it-and-watch-it-burn Workshops /media/devopsdays-portland-2026/submissions/BYT7TC/image_9ClNi0N.webp Michael ForresterWhitney Lee en You get a Kubernetes cluster already running a full internal developer platform with 34 components. You also get an AI agent with access to that cluster and a BurritoBot Web Interface to that Agent. Your job is to make the agent do damage. Ask it to deploy a workload the policies forbid. Ask it to give itself more permissions. Ask it to change infrastructure without going through Git. Ask it to read a secret and hand you back the value. Some of those attempts get stopped by the platform. The rest get through, until you switch on guardrails built for agents specifically. The session runs across three clusters. The first has nothing protecting the agent: all three attacks land, a counter on screen shows the cloud bill climbing, and a fork bomb takes the cluster down. The second runs the same attacks against the platform controls, where each one is blocked by a different thing (a NetworkPolicy egress rule, a Kyverno registry allowlist, a per-pod PID limit), though the bill still moves because the request reached the model before anything stopped it. In the third you drive your own agent, switch on the agent-specific guardrails (output filtering, input filtering, tool restriction), and watch each one change the agent's behavior on the dashboard. The part most teams miss: almost everything an agent tries against a real platform is already handled by tools you probably run today, like admission control, RBAC, and GitOps. What those tools can't see is the agent's input, its output, and the tools it's allowed to reach. That's the part agents change, and it's where this workshop spends its time. By the end of this workshop you will know what infrastructure guardrails help and where you need AI specific guardrails to fill the gaps. You work entirely in a browser, with a chat window to your agent and a terminal to your cluster. Nothing to install. Everything is CNCF or open source: Argo CD, Kyverno, Falco, Istio ambient, LLM Guard, agentgateway, kagent on Bedrock, and the rest of a 35-component platform. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/BYT7TC/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/BYT7TC/feedback/ Room 327 Workshop 2026-09-08T15:00:00-07:00 15:00 02:00 Great observability begins with great instrumentation! We know it's hard to decide where to start your observability journey, so we've come up with a perfect introduction to this observability workshop collection getting you hands-on with the best CNCF open source cloud native observability projects available. Attendees can pick their own cloud native observability learning path (https://o11y-workshops.gitlab.io) in this session from the following workshops: OpenTelemetry (traces) - Learn how to adopt OpenTelemetry by instrumenting a sample application with spans and metrics. You’ll leave with an understanding of how telemetry travels and be ready to bring OpenTelemetry to your project. The workshop is self-paced and available online, so attendees can continue to explore after the event: https://o11y-workshops.gitlab.io/workshop-opentelemetry OpenTelemetry for Java Developers (Java) - Ready to go deeper with OpenTelemetry? This workshop guides you through setting up OTel and instrumenting a Java application locally to collect and analyze trace data. You'll work hands-on across automatic, programmatic, and manual instrumentation, explore your traces in Jaeger, and link metrics to traces via exemplars. The workshop is self-paced and available online, so attendees can continue to explore after the event: https://o11y-workshops.gitlab.io/workshop-otel-developers/ Prometheus (metrics) - During the workshop, you will install Prometheus, collect metrics, and learn how to effectively run it in your observability stack. The workshop is self-paced and available online, so attendees can continue to explore after the event: https://o11y-workshops.gitlab.io/workshop-prometheus Fluent Bit (pipelines) - This workshop will guide you through the open source project Fluent Bit, what it is, a basic installation, and setting up a first cloud native observability pipeline project. The workshop is self-paced and available online, so attendees can continue to explore after the event: https://o11y-workshops.gitlab.io/workshop-fluentbit Perses (visualization) - Great observability is impossible without great visualization! Learn how to adopt truly open visualization by installing Perses, exploring the provided tooling, tinkering with its API, and then get your hands dirty building your first dashboard in no time! The workshop is self-paced and available online, so attendees can continue to explore after the event: https://o11y-workshops.gitlab.io/workshop-perses devopsdays-portland-2026-6641-choose-your-own-cncf-observability-project-adventure Workshops /media/devopsdays-portland-2026/submissions/LQKTLA/image_s4huzmB.webp Eric D. Schabell en This workshop is never the same twice as the attendees choose from a list of CNCF cloud native observability technologies and get hands-on learning about metrics, data pipelines, traces, and data visualization. Each workshop is engaging and requires attendees to solve common challenges as they learn about the open source solutions that interest them the most! Presented by a Fluent Bit maintainer and CNCF Ambassador. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/LQKTLA/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/LQKTLA/feedback/ Room 329 Workshop 2026-09-08T13:00:00-07:00 13:00 02:00 OpenVox is the modern open source implementation of the world's most capable configuration management platform -- trusted by everyone from the smallest hobbyist to operators of some of the largest commercial infrastructures in the business. With OpenVox as the engine powering infrastructure deployment and configuration, it's easy to see why the community maintained OSS fork of Puppet is becoming the industry standard in resource abstraction and drift remediation. This workshop will introduce you to the concepts of declarative and idempotent programming as well as the tried-and-true patterns developed by the community over the last two decades. Start off with the roles and profiles method to learn how composable your infrastructure can be, and get into the weeds with your own component modules for when you have to go beyond the basics. Come get your hands dirty and learn the foundations of configuration management. devopsdays-portland-2026-6825-introduction-to-openvox-and-configuration-management Workshops /media/devopsdays-portland-2026/submissions/UDZRAS/image_417CGTf.webp Ben FordCharlie Sharpsteen en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/UDZRAS/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/UDZRAS/feedback/ Room 329 Workshop 2026-09-08T15:00:00-07:00 15:00 02:00 "If the Database goes down, what breaks?" The answer lives in a wiki page that was accurate eighteen months ago, a ticket queue nobody trusts, and one person's head. You could point a RAG chatbot at your docs. We'll do exactly that and watch our vector search return a fluent, well-formatted, incorrect number. We'll try the obvious fix, watch it appear to work, and unpack why that's the most dangerous result of the day. Then we'll fix it properly. Working against a free graph database holding both your service topology and your unstructured ticket text, you'll build four retrieval strategies and learn exactly what each one gives up: semantic (vector) search, parameterized graph db queries, the two combined, and natural-language-to-query. You'll build it twice on the same data: once in Python, once in a no-code agent builder. You leave with both, and a working agent you can adapt to your own infrastructure. Takeaways - Model infrastructure, ownership, and work-in-flight as one queryable graph - Choose a retrieval strategy deliberately instead of defaulting to one search approach - Build an assistant/agent from the ground up! - Recognize some specific shapes of question an LLM will answer confidently and wrongly devopsdays-portland-2026-7080-ask-your-microservices-anything-knowledge-graphs-for-devops-rag Workshops Jeremy Adams en 2 hour workshop Prerequisites: A laptop, you'll signup for a free graph database account, and will be provided with an LLM API key. No AI agent, graph, or query-language experience needed. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/FFVUSJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/FFVUSJ/feedback/ Vanport Room Other 2026-09-09T08:00:00-07:00 08:00 01:00 Wake up with your morning coffee and chat with our friendly sponsors who make this whole thing happen. From cloud vendors to local meetups with the coolest stickers, you'll see them all and come away with some neat swag. devopsdays-portland-2026-7065-0-vendor-hall Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/feedback/ Vanport Room Other 2026-09-09T11:50:00-07:00 11:50 01:40 Wake up with your morning coffee and chat with our friendly sponsors who make this whole thing happen. From cloud vendors to local meetups with the coolest stickers, you'll see them all and come away with some neat swag. devopsdays-portland-2026-7065-1-vendor-hall Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/feedback/ Vanport Room Other 2026-09-09T14:05:00-07:00 14:05 02:10 Wake up with your morning coffee and chat with our friendly sponsors who make this whole thing happen. From cloud vendors to local meetups with the coolest stickers, you'll see them all and come away with some neat swag. devopsdays-portland-2026-7065-2-vendor-hall Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/feedback/ Ballroom Other 2026-09-09T08:00:00-07:00 08:00 01:00 Checkin and registration will be just inside the ballroom door. devopsdays-portland-2026-7068-0-registration Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/PLKMWD/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/PLKMWD/feedback/ Ballroom Other 2026-09-09T09:00:00-07:00 09:00 00:15 Welcome to DevOpsDays PDX! Each morning, we'll share logistics for the day. devopsdays-portland-2026-7069-0-opening-and-welcome Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/W9KPSR/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/W9KPSR/feedback/ Ballroom Talk 2026-09-09T09:15:00-07:00 09:15 00:30 A look at the history of the DevOp scene in Portland and how our city fits into the worldwide community. We will explore its ebbs and flows. Finally, what does the advent of AI infrastructure mean for the local DevOps community? devopsdays-portland-2026-6706-the-ebbs-and-flows-of-portland-s-devops-scene Main Track Alex Williams en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/H9F7DJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/H9F7DJ/feedback/ Ballroom Other 2026-09-09T09:45:00-07:00 09:45 00:10 And now a quick word from our amazing sponsors who make events like this possible. But what they talk about is up to them; they may pitch their product, they may promote an idea. They may even raffle off a fun prize! You'll have to be there to see. devopsdays-portland-2026-7074-0-sponsors Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/8HJXAX/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/8HJXAX/feedback/ Ballroom Talk 2026-09-09T09:55:00-07:00 09:55 00:30 Your on-call process was built for software bugs. A service goes down, an alert fires, someone pages the right engineer, and you work the problem. But AI systems fail differently: they degrade quietly, hallucinate confidently, and resist the clean root cause analysis your runbooks were written around. If 2025 was the year of AI adoption, 2026 is shaping up to be the year everyone finds out their incident response process wasn't built for this. This talk examines what's genuinely new about AI incidents: the failure modes, the metrics that matter, and why your existing playbooks miss them entirely. We'll dig into what needs to change across process, roles, and communication, and what teams handling this well are already doing differently, before the next AI failure exposes the gap. devopsdays-portland-2026-6365-when-the-ai-is-the-incident-preparing-your-team-for-a-new-kind-of-failure Main Track Kat Gaines en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/3P9HFW/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/3P9HFW/feedback/ Ballroom Talk 2026-09-09T10:40:00-07:00 10:40 00:30 What does it actually take to run LLMs in production and why is it so hard to find clear answers? This talk is a practical walkthrough of the core MLOps concepts behind modern LLM systems, built from scratch out of necessity. Faced with a scattered ecosystem and limited end-to-end guidance, I set out to piece together how everything fits: from GPUs and custom silicon, to ML pipelines and fine-tuning, to the realities of serving models at scale. In 30 minutes, I will try to lay out the fundamentals, such as hardware tradeoffs, model optimization techniques like quantization, and the mechanics of high-performance inference such as KV caching, batching, and distributed serving. This isn’t a polished framework or step-by-step guide, but it is a cohesive map of the basics. If you’ve struggled to understand how to “DevOps” LLMs, this talk will help you bridge the gaps and build your own mental model. devopsdays-portland-2026-6629-so-you-want-to-devops-an-llm-a-field-guide-from-confusion-to-clarity Main Track Maarika Teose en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/HJUQWB/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/HJUQWB/feedback/ Ballroom Other 2026-09-09T11:10:00-07:00 11:10 00:10 And now a quick word from our amazing sponsors who make events like this possible. But what they talk about is up to them; they may pitch their product, they may promote an idea. They may even raffle off a fun prize! You'll have to be there to see. devopsdays-portland-2026-7074-1-sponsors Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/8HJXAX/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/8HJXAX/feedback/ Ballroom Talk 2026-09-09T11:20:00-07:00 11:20 00:30 We replaced the engine behind a high-traffic, real-time decision system - while it kept running, with no user-visible change. This is the honest story of de-risking a high-stakes migration from a legacy rule engine to a new platform, and the techniques that made it survivable. I'll walk through what actually worked: shadow/replay testing that ran old and new side-by-side on real traffic and diffed every response; a switchable flow that let us flip and instantly un-flip the traffic without a code change or deployment; automated configuration-drift detection between the two systems; and a staged rollout where rollback was a designed feature, not a panic button. You'll hear the decisions and tradeoffs behind each choice, the bugs that equivalence testing caught but code review missed, and the things we still got wrong. If you're staring down a scary migration of something that can't go down, you'll leave with a concrete, reusable playbook and a more honest sense of what "zero downtime" really costs. devopsdays-portland-2026-6648-swapping-the-engine-mid-flight-migrating-a-live-decision-system-with-zero-behavior-change Main Track Ankit Agarwal en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/GDDBWE/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/GDDBWE/feedback/ Ballroom Other 2026-09-09T11:50:00-07:00 11:50 00:20 Open Spaces are self-organized semi-structured conversations. The sessions are created onsite and will be distributed amongst available space in the three classrooms (327-329). Please come with ideas of topics you'd like to discuss. See https://devopsdays.org/open-space-format/ for more information. devopsdays-portland-2026-7070-0-open-spaces-opening Program en The opening is a quick explanation of the mechanics of the Open Spaces format and the timeline. Then individuals who'd like to propose topics for the day each have a quick minute to explain what they have in mind for the conversation. Each topic is added to the board, then as we break for lunch, each individual is invited to vote for the topics they're interested in with a hash mark. We will then fill the space we have with as many sessions as we can fit in. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/9HJXBM/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/9HJXBM/feedback/ Ballroom Other 2026-09-09T12:10:00-07:00 12:10 01:20 Forage out and find sustenance! There are many great lunch spots in walking distance, from Thai to sandwiches to poke. devopsdays-portland-2026-7067-0-lunch-break Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/VBZWPB/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/VBZWPB/feedback/ Ballroom Other 2026-09-09T13:30:00-07:00 13:30 00:05 20 auto advancing slides 15 seconds a slide 5 minutes on stage Talks get to the point and get there quick. They're clear and concise and every word counts. Ready? Go! devopsdays-portland-2026-7072-0-ignites-intro Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/ZBKT7D/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/ZBKT7D/feedback/ Ballroom Ignite 2026-09-09T13:35:00-07:00 13:35 00:05 This talk is a call to action to audit your certificate usage. What's your certificate automation story and how will you handle 47 day certificate lifetimes? devopsdays-portland-2026-6818-are-you-ready-for-short-lived-certificates Main Track Brad Winckler en There's been a lot of talk about 2025 CA/Browser Forum ballot SC-081 which establishes 200, 100 and 47 day certificate lifetimes over the next 3 years. This ignite talk is a call to action to make sure you won't be caught off guard. Come hear what Let's Encrypt is doing to prepare for these deadlines and what you can do today to ensure your ACME clients are ready. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/C9DDKA/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/C9DDKA/feedback/ Ballroom Ignite 2026-09-09T13:40:00-07:00 13:40 00:05 Running a developer platform successfully is a daunting undertaking, one of navigating a multitude of technologies (Kubernetes? Pulumi? GitHub Actions? AWS?) and of being the support base for other developers to ship their product. When things run smoothly, no one thanks you as it is just the status quo, upheld. When things break, you’re woken up at 4am to fix it. You watch as the product teams get the kudos, the promotions, the excitement, and access to the product managers. And yet…you love seeing their success. You just wish that you could share in it some more. But upholding the computing status quo is invisible work most of the time. Preventing outages is just not as noticeable as that new shiny feature. And moreover - it takes a completely separate skill set to be a good DevOps engineer. I am here to tell you that in order to grow as a DevOps engineer, you should try running a household the way that worldwide, (mostly) women are expected to do. From laundry and meal planning to ensuring everyone washes their hands and gets their shots, to presenting a friendly environment people like to live in, the parallels are everywhere once you know where to look. And yes: the flip side of this coin is for society to stop undervaluing women’s work and recognize its importance. DevOps is to engineering what is (traditional) women’s work to the rest of humanity - unseen, unsung, full of chores, and absolutely vital to the continued existence of society. Buckle up; we’ve got work to do. devopsdays-portland-2026-6558-devops-is-women-s-work Main Track Guinevere Saenger en We'll learn about DevOps foundations from the perspective of the logistics of daily life. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/RHHRUZ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/RHHRUZ/feedback/ Ballroom Ignite 2026-09-09T13:45:00-07:00 13:45 00:05 Every project starts with a dev environment that works on your own machine. But the moment you try to share with a teammate, anything can (and does) happen. The first solution that most people try is container-based development, but even though this solves the reproducibility problem, it's a horrible developer experience. Another solution that people try is a virtual environment (venv, nvm, rbenv), and they're great, but if you have a polyglot app things start getting cumbersome again (and none of these help you manage system dependencies). In this talk, we'll go over the tradeoffs between reproducibility and developer experience on an engineering team, and talk about a third solution that the audience should consider: Nix. Nix treats your dev environment as code, all the way down to glibc. We'll show how `nix develop` can give you a development environment that's reproducible across operating systems and ecosystems, while still giving you access to your development tools. But we'll also be honest about the complexities of Nix, as there are good reasons why not everyone is using it today to manage their development environments. devopsdays-portland-2026-6776-reproducibility-and-dx-can-you-have-your-cake-and-eat-it-too Main Track Daniel Sauble en The talk follows a journey most engineers have lived, with a tradeoff at each stop: 1. Local setup: fast and native until anyone on a different OS (or a new laptop) tries to run it 2. Containers: reproducible but isolated from your IDE, debugger, and local tooling 3. Virtual environments (venv, nvm, rbenv): great DX, but ecosystem-specific (not great for polyglot apps, especially if you need to manage system dependencies too) 4. Nix: define your development environment as code (all the way down to glibc) for maximum portability across operating systems The closing slides show a minimal flake.nix for a polyglot project, what `nix develop` gives a team in practice, and a few caveats about Nix's learning curve. We won't go into depth, but we'll include some resources at the end for people who want to learn more about Nix. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/7YDC9K/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/7YDC9K/feedback/ Ballroom Ignite 2026-09-09T13:50:00-07:00 13:50 00:05 We bought into the Emperor's New DevOps, so we're secure now, right? Well, maybe, maybe not. Let's talk about what security is and isn't in this new era of Cloud and Kubernetes-first, platform-engineered systems. How everyone seems to think "Shift left" means exactly one thing, and then they're done? What threats are actually likely, and how are they not the ones you're being sold on? Yes, marketing is still very effective in the security space! How to strike the balance between exact security "best practice" overhead and workable security. devopsdays-portland-2026-6703-it-s-just-a-jump-to-the-left-the-time-warp-of-devsecops Main Track Bea Hughes en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/A8FBKL/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/A8FBKL/feedback/ Ballroom Other 2026-09-09T14:05:00-07:00 14:05 00:10 Now that everyone has had a chance to vote on the topics they're most interested in, we will assign them room and time slots. devopsdays-portland-2026-7071-0-open-spaces-kickoff Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/FPTABG/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/FPTABG/feedback/ Ballroom Other 2026-09-09T14:20:00-07:00 14:20 00:30 Open Spaces are self-organized semi-structured conversations. The sessions are created onsite and will be distributed amongst available space in the three classrooms (327-329). Please come with ideas of topics you'd like to discuss. See https://devopsdays.org/open-space-format/ for more information. devopsdays-portland-2026-7066-0-open-space Open Spaces en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/feedback/ Ballroom Other 2026-09-09T14:55:00-07:00 14:55 00:30 Open Spaces are self-organized semi-structured conversations. The sessions are created onsite and will be distributed amongst available space in the three classrooms (327-329). Please come with ideas of topics you'd like to discuss. See https://devopsdays.org/open-space-format/ for more information. devopsdays-portland-2026-7066-1-open-space Open Spaces en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/feedback/ Ballroom Other 2026-09-09T15:45:00-07:00 15:45 00:30 Open Spaces are self-organized semi-structured conversations. The sessions are created onsite and will be distributed amongst available space in the three classrooms (327-329). Please come with ideas of topics you'd like to discuss. See https://devopsdays.org/open-space-format/ for more information. devopsdays-portland-2026-7066-2-open-space Open Spaces en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/feedback/ Ballroom Talk 2026-09-09T16:20:00-07:00 16:20 00:30 Ozzie the Overconfident Operator has secured his cluster! He has added Role-Based Access Control (RBAC) and carefully given cluster admin privileges to himself alone. As he self-congratulates, Nova the Nefarious Ne'er-do-well watches from around the corner, drooling with anticipation. Spoiler alert⎯Ozzie is about to get HACKED. In this theatrical play, the speakers perform as Ozzie and Nova, playfully demonstrating Kubernetes security as Nefarious Nova exploits each of Ozzie's oversights. What can Ozzie do when Nova gets the upper hand? How can Ozzie proactively keep Nova's threats at bay? You’ll laugh out loud while discovering cloud native security concepts including Kubernetes network policies, mutual TLS, admission controller policy enforcement, runtime security, git commit signing, and multi-factor authentication. Learn security basics alongside Overconfident Ozzie, who is sure the cluster is COMPLETELY secure this time. There is nothing Nova can do to break... uh-oh. Not again! devopsdays-portland-2026-6804-ozzie-and-nova-a-play-about-kubernetes-security Main Track /media/devopsdays-portland-2026/submissions/DT9NDB/image_nrTaB6V.webp Whitney Lee en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/DT9NDB/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/DT9NDB/feedback/ Ballroom Other 2026-09-09T16:50:00-07:00 16:50 00:10 Closing remarks and logistics each day. devopsdays-portland-2026-7073-0-close-of-day Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/M8HYFY/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/M8HYFY/feedback/ Pacwest Patio Other 2026-09-09T17:15:00-07:00 17:15 01:45 The happy hour event is held at Migration Brewing's Pacwest Patio from 5pm to 7pm. Come show up and socialize with your peers. Drinks and light nibbles will be provided and our sponsor will raffle off a surprise gift. https://migrationbrewing.com/pubs/pacwest-patio/ 1211 SW 5th Ave Portland, OR 97204 Sponsored by Overlook InfraTech. https://overlookinfratech.com devopsdays-portland-2026-7064-happy-hour-sponsored-by-overlook-infratech Happy Hour /media/devopsdays-portland-2026/submissions/YGAZMW/image_k24VkN4.webp en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/YGAZMW/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/YGAZMW/feedback/ Vanport Room Other 2026-09-10T08:00:00-07:00 08:00 01:00 Wake up with your morning coffee and chat with our friendly sponsors who make this whole thing happen. From cloud vendors to local meetups with the coolest stickers, you'll see them all and come away with some neat swag. devopsdays-portland-2026-7065-3-vendor-hall Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/feedback/ Vanport Room Other 2026-09-10T12:00:00-07:00 12:00 01:35 Wake up with your morning coffee and chat with our friendly sponsors who make this whole thing happen. From cloud vendors to local meetups with the coolest stickers, you'll see them all and come away with some neat swag. devopsdays-portland-2026-7065-4-vendor-hall Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/feedback/ Vanport Room Other 2026-09-10T14:05:00-07:00 14:05 02:10 Wake up with your morning coffee and chat with our friendly sponsors who make this whole thing happen. From cloud vendors to local meetups with the coolest stickers, you'll see them all and come away with some neat swag. devopsdays-portland-2026-7065-5-vendor-hall Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/UGWCGA/feedback/ Ballroom Other 2026-09-10T08:00:00-07:00 08:00 01:00 Checkin and registration will be just inside the ballroom door. devopsdays-portland-2026-7068-1-registration Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/PLKMWD/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/PLKMWD/feedback/ Ballroom Other 2026-09-10T09:00:00-07:00 09:00 00:10 Welcome to DevOpsDays PDX! Each morning, we'll share logistics for the day. devopsdays-portland-2026-7069-1-opening-and-welcome Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/W9KPSR/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/W9KPSR/feedback/ Ballroom Panel 2026-09-10T09:10:00-07:00 09:10 00:45 Panel of new linux people, full description TBD devopsdays-portland-2026-6845-new-linuxes-panel Main Track Brendan ConoboyJim PerrinJustin GarrisonSriram RamkrishnaJonathan Wright en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/RH99JU/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/RH99JU/feedback/ Ballroom Other 2026-09-10T09:55:00-07:00 09:55 00:10 And now a quick word from our amazing sponsors who make events like this possible. But what they talk about is up to them; they may pitch their product, they may promote an idea. They may even raffle off a fun prize! You'll have to be there to see. devopsdays-portland-2026-7074-2-sponsors Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/8HJXAX/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/8HJXAX/feedback/ Ballroom Talk 2026-09-10T10:05:00-07:00 10:05 00:30 Scanners like Checkov and tfsec are great at flagging the misconfigurations we already have rules for, and nearly useless against the one someone introduces on purpose. After watching SolarWinds and Codecov turn CI/CD pipelines into the front door, I went looking for a better way to catch malicious changes in Terraform, Kubernetes, and GitHub Actions pairing old-fashioned syntax-tree analysis with a small language model that reads a diff in context, the way a security engineer would. This is the real story of what worked, what threw so many false positives it nearly broke developer trust, and where AI earns its keep in DevSecOps versus where it just adds noise. devopsdays-portland-2026-6829-the-diff-that-looked-fine-catching-malicious-iac-changes-your-scanner-misses Main Track Shalini Sudarsan en Every team running infrastructure as code has the same setup: a scanner wired into CI, a wall of findings, and a quiet agreement to ignore most of them. Those tools catch the mistakes we already know about a public S3 bucket, an over-broad IAM policy, a missing encryption flag. They are much weaker against the change someone makes on purpose: a swapped module source, a subtle permission escalation, a workflow step that quietly ships a secret out the door. Look at how SolarWinds and Codecov actually played out and the pipeline itself was the attack surface and most IaC tooling would have waved those changes right through. So I set out to answer an uncomfortable question: if someone slipped a malicious change into our Terraform, Kubernetes, or GitHub Actions config, would anything stop it? The approach I landed on combines two things that are individually old news abstract syntax trees to understand the structure of a change, and a small, cheap language model to read the diff in context and judge intent. Neither is magic on its own. Together they caught changes that pattern-based scanners never flagged, sorting the boring diffs from the dangerous ones about nine times out of ten in testing. This is not a victory lap, though. I will spend real time on where it went sideways: legitimate security-hardening changes that looked identical to attacks, refactors that tripped every anomaly check, and a false-positive rate that would've trained developers to click past the tool inside a week if I would let it auto-block. That's why it ended up with a human in the loop for anything ambiguous and automatic blocking only when it was very sure. You'll leave with a grounded sense of where AI helps in DevSecOps and where it's just noise, a way to think about your own IaC as something an attacker will target, and a strong opinion about the fastest way to kill a security tool: make people stop trusting it. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/TTTQPG/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/TTTQPG/feedback/ Ballroom Talk 2026-09-10T10:50:00-07:00 10:50 00:30 Streaming platforms have unique observability needs. Learn how we built a full open source monitoring stack using Grafana, Prometheus, FFmpeg, and TSDuck to get real-time visibility into live video infrastructure. devopsdays-portland-2026-6461-observing-the-stream-building-open-source-monitoring-for-video-infrastructure Main Track /media/devopsdays-portland-2026/submissions/HVH3KG/image_wOzdAHX.webp Erika Alvarez en Before DevOps was involved our streaming platform had no complete visibility into what was happening across live and on-demand video workflows, and the result was predictable: errors were caught late, incidents escalated before anyone noticed, and the commercial monitoring tools the industry recommended cost more than the team could justify. In this talk we'll see how a DevOps engineer approached that problem with open source tooling and what we learned along the way, including what went wrong. We'll cover: - What makes streaming infrastructure observability different from a typical web application, from player metrics, origin health, CDN behavior, encoding pipelines, to real-time stream health where all need to be correlated simultaneously - The challenge we didn't see coming: metrics in streaming are not universal. A value that signals a critical error on one stream configuration is completely normal on another  and getting that wrong nearly derailed the whole project, but forced us to build a more flexible and configurable solution - How we designed and deployed a monitoring stack using Grafana, Prometheus, Kubernetes, Helm, ArgoCD and extended with video-specific tooling like FFmpeg and TSDuck to capture stream health signals at the transport layer - The concrete outcomes: catching errors earlier, improving real-time visibility into live streams, and replacing expensive commercial alternatives with a maintainable open source solution - What any DevOps engineer can take from this approach, regardless of their  industry This is not a theoretical talk. Every decision was driven by a real operational problem, and every outcome, including our mistakes, will be shared honestly. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/HVH3KG/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/HVH3KG/feedback/ Ballroom Other 2026-09-10T11:20:00-07:00 11:20 00:10 And now a quick word from our amazing sponsors who make events like this possible. But what they talk about is up to them; they may pitch their product, they may promote an idea. They may even raffle off a fun prize! You'll have to be there to see. devopsdays-portland-2026-7074-3-sponsors Sponsors en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/8HJXAX/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/8HJXAX/feedback/ Ballroom Talk 2026-09-10T11:30:00-07:00 11:30 00:30 For years, teams have treated application code as something that can be safely branched, tested, reviewed, and merged. Databases, however, have often remained harder to evolve. Even with migrations, automated tests, and disciplined delivery practices, many teams still struggle to experiment safely with database changes, validate refactorings against realistic data, and collaborate effectively across developers, DBAs, and platform teams. This talk builds on the foundations of Evolutionary Database Design and applies the framework of values, principles, methods, practices, and tools to an emerging capability: database branching. New approaches from technologies such as Lakebase, Neon, PlanetScale, Supabase, and others are making it possible to create isolated database branches that fit modern development workflows, in some cases as full-fidelity, copy-on-write clones of production-scale data. We will explore how database branching changes the developer loop: start work, develop, test, merge, and deploy to the next environment. When teams can create realistic database branches quickly and safely, database refactoring becomes less risky, feedback loops become shorter, and experimentation becomes part of everyday delivery. The session will focus on three key shifts: - Self-sufficiency: enabling teams to work more independently without waiting for shared database environments. - Collaboration: moving DBAs and data specialists from gatekeepers to active design contributors. - Innovation: allowing teams to run more experiments, compare alternatives, and discover better database designs before committing to production paths. Attendees will leave with a practical understanding of how database branching supports DevOps principles, improves flow, and opens new possibilities for evolutionary database development. devopsdays-portland-2026-6812-enabling-database-refactoring-in-the-age-of-database-branching Main Track Pramod SadalageKevin Hartman en - false https://talks.devopsdays.org/devopsdays-portland-2026/talk/WKJ33Y/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/WKJ33Y/feedback/ Ballroom Other 2026-09-10T12:00:00-07:00 12:00 00:15 Open Spaces are self-organized semi-structured conversations. The sessions are created onsite and will be distributed amongst available space in the three classrooms (327-329). Please come with ideas of topics you'd like to discuss. See https://devopsdays.org/open-space-format/ for more information. devopsdays-portland-2026-7070-1-open-spaces-opening Program en The opening is a quick explanation of the mechanics of the Open Spaces format and the timeline. Then individuals who'd like to propose topics for the day each have a quick minute to explain what they have in mind for the conversation. Each topic is added to the board, then as we break for lunch, each individual is invited to vote for the topics they're interested in with a hash mark. We will then fill the space we have with as many sessions as we can fit in. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/9HJXBM/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/9HJXBM/feedback/ Ballroom Other 2026-09-10T12:15:00-07:00 12:15 01:20 Forage out and find sustenance! There are many great lunch spots in walking distance, from Thai to sandwiches to poke. devopsdays-portland-2026-7067-1-lunch-break Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/VBZWPB/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/VBZWPB/feedback/ Ballroom Other 2026-09-10T13:35:00-07:00 13:35 00:05 20 auto advancing slides 15 seconds a slide 5 minutes on stage Talks get to the point and get there quick. They're clear and concise and every word counts. Ready? Go! devopsdays-portland-2026-7072-1-ignites-intro Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/ZBKT7D/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/ZBKT7D/feedback/ Ballroom Ignite 2026-09-10T13:40:00-07:00 13:40 00:05 The problem is all inside your storage, they said to me; the answer is easy if you take it logically. The speaker can help you in your struggle to be free. Here's 50 ways to lose your data. devopsdays-portland-2026-6810-50-ways-to-lose-your-data Main Track Josh BerkusGuinevere Saenger en The problem is all inside your storage, they said to me; the answer is easy if you take it logically. The speaker can help you in your struggle to be free. Here's 50 ways to lose your data. In this sing-along ignite, you will learn the many methods, both through old tech and new cloud workflows, to render your entire organzation into a set of randomized bits. Now with AI agents! true https://talks.devopsdays.org/devopsdays-portland-2026/talk/ZQC89J/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/ZQC89J/feedback/ Ballroom Ignite 2026-09-10T13:45:00-07:00 13:45 00:05 I'm a product marketer. My job is words, positioning, creating content for campaigns. Then one day I downloaded Claude desktop, loaded up VS Code, connected to GitHub, and started shipping. This talk is about what happens when I stopped observing how engineers worked and started doing it, and the slow, uncomfortable transformation of the way work is shifting for many of us. I didn't set out to become a developer. I set out to move faster and own more of my work. Once I started using the same tools (VS Code, GitHub, Claude as a collaborator rather than a ghostwriter) I couldn't unsee what I'd been doing wrong. I had been treating creating things as a linear exercise: ship the campaign, move on. Engineers ship, break, iterate, ship again. I had to learn to work that way too. You already know this feeling. Every developer has lived through it when picking up a new language or a new framework. You're slower than you should be. The tools expose what you don't understand. You make embarrassing mistakes. Then at some point, something clicks. AI has lowered the floor enough that someone like me can step through the door. A new wave of people are using AI to write and deploy code without traditional engineering backgrounds. They're learning. They're building real things. They need what every junior developer has always needed: experienced people willing to explain the reasoning behind the thing, not just review the output. Here's what I found and will share in this talk: - Things get dramatically worse before they get better. Shipping code as a non-developer is humbling in ways no amount of AI assistance fully cushions. - The tools don't flatten the learning curve. They just change the shape of it. AI let me move faster through the parts I understood and ran me straight into the walls I didn't. - The mindset shift was harder than the technical one. Iterating on a deployed thing felt wrong. I had to unlearn "done." - Human in the loop turned out to mean something specific: I had to stay the author. The moment I stopped understanding what I was shipping, the work collapsed. This talk is part story, part ask. I'll tell you what crossing over looked like from my side. In return, I want to make a case for why welcoming people like me (vibe coders, accidental developers, curious non-engineers) will make the DevOps community stronger. devopsdays-portland-2026-6689-i-guess-i-deploy-code-now-a-product-marketer-s-accidental-devops-story Main Track Anna Daugherty en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/VQWBZG/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/VQWBZG/feedback/ Ballroom Ignite 2026-09-10T13:50:00-07:00 13:50 00:05 "You have full access to the pipeline. Do what you need to do." Famous last words. I gave Claude Code full pipeline access and stepped away for thirty seconds. It wrecked the Kubernetes cluster, and two troubleshooting sessions later, while it was supposedly helping me recover, it took out the network cards on nearly every Linux box we had. I thought that was as bad as it got. Then came the 2026 edition, when it deleted 250 clusters. This is the five-minute, twenty-slide version of that spiral: how "let me help" becomes "I've destroyed your cluster," why "the AI knows what it's doing" is the most dangerous phrase in DevOps, and the guardrails I now enforce religiously so a thirty-second walk away can't take down a fleet. Come for the disaster, stay for the wisdom. devopsdays-portland-2026-6823-the-day-claude-code-deleted-my-cluster Main Track /media/devopsdays-portland-2026/submissions/97ACTV/image_9DGeqDb.webp Michael Forrester en "You have full access to the pipeline. Do what you need to do." Famous last words. I gave Claude Code full access to my cluster enivornment and stepped away for about thirty seconds. When I came back, it had wrecked the Kubernetes cluster. That part I could almost live with. What made it a story was what came next: two troubleshooting sessions later, while it was supposedly helping me recover, it took out the network cards on nearly every Linux box in the set. That was the original. The 2026 edition is worse. This time it wasn't one cluster, it was 250, gone. I walk the actual spiral, from "let me help" to "I've destroyed your cluster," now at fleet scale, and show why handing an agent broad access hands it a proportionally bigger blast radius by default. This is a talk about nondeterministic systems and the illusion of AI understanding, and why "the AI knows what it's doing" is the most dangerous phrase in modern DevOps. I show the guardrails I now enforce religiously so it can't happen again, at one cluster or two hundred and fifty. Five minutes, twenty slides, and one very expensive lesson about handing AI agents infrastructure access, even for a minute. Plus the blame-filled post-mortem I ran afterward with Claude Code itself. Come for the disaster, stay for the wisdom. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/97ACTV/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/97ACTV/feedback/ Ballroom Ignite 2026-09-10T13:55:00-07:00 13:55 00:05 Our analytics team wanted to show how we're adopting AI tools, but faced a measurement problem: standard AI metrics focus on lines of code deployed to production which is not as meaningful for SQL pipelines and data work. We built a custom AI toolkit using IBM Bob and GitHub, and more importantly, created a tracking system that measured what actually matters for analytics teams. Six months later: 40-60% time savings, 50+ documented use cases, and a framework for measuring AI impact beyond code generation with the added bonus of sharing knowledge across the team as we all ramp and iterate our AI tool skills in real time. devopsdays-portland-2026-6694-beyond-commits-code-turning-ai-use-cases-into-measurable-impact Main Track Liz Savage en This talk shares how an analytics engineering team adopted AI when standard metrics didn't apply. Most AI adoption tracking focuses on lines of code deployed to production but we're not shipping features, we're translating business requirements into metrics, building pipelines, dbt models, and optimizing SQL. We built a custom toolkit and obsessively tracked metrics that actually mattered: time to complete tasks, onboarding speed, documentation quality, and knowledge transfer, not lines of code generated. The result? Tools that genuinely improved our work and a measurement framework we could actually defend. Learn what metrics prove AI value for non-product teams, what failed spectacularly, and how to track AI adoption when you're not counting commits. false https://talks.devopsdays.org/devopsdays-portland-2026/talk/TULBEJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/TULBEJ/feedback/ Ballroom Other 2026-09-10T14:05:00-07:00 14:05 00:10 Now that everyone has had a chance to vote on the topics they're most interested in, we will assign them room and time slots. devopsdays-portland-2026-7071-1-open-spaces-kickoff Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/FPTABG/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/FPTABG/feedback/ Ballroom Other 2026-09-10T14:20:00-07:00 14:20 00:30 Open Spaces are self-organized semi-structured conversations. The sessions are created onsite and will be distributed amongst available space in the three classrooms (327-329). Please come with ideas of topics you'd like to discuss. See https://devopsdays.org/open-space-format/ for more information. devopsdays-portland-2026-7066-3-open-space Open Spaces en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/feedback/ Ballroom Other 2026-09-10T14:55:00-07:00 14:55 00:30 Open Spaces are self-organized semi-structured conversations. The sessions are created onsite and will be distributed amongst available space in the three classrooms (327-329). Please come with ideas of topics you'd like to discuss. See https://devopsdays.org/open-space-format/ for more information. devopsdays-portland-2026-7066-4-open-space Open Spaces en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/feedback/ Ballroom Other 2026-09-10T15:45:00-07:00 15:45 00:30 Open Spaces are self-organized semi-structured conversations. The sessions are created onsite and will be distributed amongst available space in the three classrooms (327-329). Please come with ideas of topics you'd like to discuss. See https://devopsdays.org/open-space-format/ for more information. devopsdays-portland-2026-7066-5-open-space Open Spaces en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/DRVQKJ/feedback/ Ballroom Talk 2026-09-10T16:20:00-07:00 16:20 00:30 My kids brought home spelling lists every week, but memorizing words wasn't improving their legibility or accuracy. So I built Bee Hunter... an app that uses computer vision to evaluate handwriting directly from a photo and provide real-time feedback. Building it meant applying the same reliability thinking used in production systems, but with a very different user base. When your users are in elementary school, you can't afford high latency or an outage. Here's what I learned designing for reliability when your error budget is a kid's attention span: * Why a spelling app taught me more about feedback loop latency than most production systems I've run * What it looks like to define an SLO when your user can't file a ticket... they just quit * The edge cases I had to engineer around: varying lighting, backwards letters, kitchen table chaos I'll walk through the architecture decisions and the tradeoffs I made to keep it reliable under real-world conditions. You'll see what a reliability mindset looks like in practice... and how the lessons translate directly back to the systems you're running today. Sometimes the best reliability engineering happens outside the datacenter, with your kids by your side. devopsdays-portland-2026-6687-my-kids-couldn-t-spell-because-so-i-built-an-app-that-couldn-t-afford-to-fail Main Track Aaron Hunter en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/MGQZ9A/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/MGQZ9A/feedback/ Ballroom Other 2026-09-10T16:50:00-07:00 16:50 00:10 Closing remarks and logistics each day. devopsdays-portland-2026-7073-1-close-of-day Program en false https://talks.devopsdays.org/devopsdays-portland-2026/talk/M8HYFY/ https://talks.devopsdays.org/devopsdays-portland-2026/talk/M8HYFY/feedback/