The Platform Engineering Journey en

People in tech usually don't struggle with the technical implementation, but they often struggle with technical decisions and strategy. Over the years, I have observed trends come and go, but there have been a couple that truly make a difference. One of the more recent trends is that of Platform Engineering, which is the accumulation of what we have learned over the years in IT-operations, combining DevOps with a product mindset to make sure that developers can focus on creating business value instead of having to deal with infrastructure and cloud architecture decisions.

In this talk, I would like to explore some of the common pitfalls, talk about my experiences going from traditional sysadmin to platform engineer, and give you all a great start on your platform engineering journey.

The Agony of Choice – Build, Buy, or Lease? en no

Platform strategy, it's simple! Or, perhaps not? After all, there are tons of ready-made solutions, open-source projects, and not least, opinions out there.

Many believe a platform must be built from scratch, some think you should buy as much as possible as services, and some just want to show up to a table already set. There are pros and cons to all of these, and when we also have an industry in constant motion – it quickly becomes difficult to hit the mark.

The choice between building, buying, or leasing is rarely simple, but it is a crucial strategic decision. There is no one-size-fits-all answer, and it's not always the case that you only have one platform to make the choice for.

But what are the advantages and disadvantages of the different approaches, and in what direction are the trends heading these days? Is it really that expensive to build, and how can one avoid buying a pig in a poke? This presentation aims to shed light on precisely these questions, and give you some concrete perspectives for navigating the choices and avoiding the most common pitfalls.

Scaling GitHub Actions with Kubernetes en

While public runners are an efficient way of getting your GitHub workflows up and running, there are scenarios where you cannot use them. Be it regulatory, security, or just performance, there is a fair chance that you have to deploy your own actions runners at some point.

One of the main benefits of Kubernetes is that it scales, the drawback is that it can be a pain to configuration. But we in the cloud native community are masters at abstractions, and by adding controllers or operators we can easily automate away the pain.

Let us take a good look at the Actions Runner Controller from GitHub, an efficient way of deploying self-hosted runners.

"I choo-choo-choose you"; or how to build modern platforms in a 200 year old industry. en

Train companies know the importance of platforms and infrastructure, but what do we focus on while building platform services for critical systems that so many people rely on every day?

At BaneNOR, we have all the bells and whistles, and platforms of all sorts. Application, data, integration, and everything from state of the art technology, to legacy systems. How this is structured is a continuous work in progress and evolves in tandem with what the community discovers. Every day try to give developers a good place to run applications, while keeping stakeholders up to date, while keeping everything secure and compliant.

In this presentation we want to go through some of our strategic technical and sociotechnical choices, as well as pain points, pitfalls and low-hanging fruits. All on board the Platform Engineering express train!

The art of doing code commits en

Git commits are so simple that it is easy to do it bad. When you try to figure out a codebase, it really doesn't help that you have a Git history filled with messages like "testing", "fix bug", or even obscure messages where you just mashed your hands on your keyboard to type out "asdasd".

What if I told you that others struggled with this enough to create a specification for commit messages? A system to the madness, so to speak. A tool that promotes better readability, makes it easier to collaborate, and enables automation of all those boring chores you have to do before release.

In this talk we will be looking at Conventional Commits. It is easy to adopt, simple to implement, and powerful in its impact. Say goodbye to "mystery meat" commit messages, and welcome a system that works for you, not against you. You'll thank me later.

From null to applications on Kubernetes en

For many, Kubernetes is a big hurdle to get over. In actuality, it's just a platform for running applications with many built-in features, and is no less complicated than running applications on any other platform. There is a lot that happens up to the point where your code hits the node, and even after that one has to be weary that defaults usually aren't the best approach.

In this session, we start with an empty git repository, build a shiny new application, squeeze it into a container which we then ship to a Kubernetes cluster, and in the end expose it to the world. Think of this as a development cycle speed run.

This talk aims to go through the process of developing an application for Kubernetes. We start by writing our application, we run tests and package it into a container, and then go through all the caveats of running our application with Kubernetes. This is mostly a hands-on presentation, with few slides.

The app is developed in a programming language that fits the conference, or Golang by default.

Building an Internal Developer Platform on Azure en

As the operations team strive towards DevOps, some major shifts in responsibility are happening. We no longer want to have dedicated teams for operations, but give that responsibility to those who develop the product or service. But to ensure that the cloud infrastructure is set up in a secure and efficient manner, we need to help bridge the gap between developers and cloud provider. For this talk, I want to explain what an Internal Developer Platform is, and how we can build one using cloud native tools available on Microsoft Azure.

Workshop - GitOps with Flux en

In the cloud native era, GitOps have become one of the preferred methods of handling application deployment. When developers have created their applications, we don’t want to write imperative and bespoke pipelines for deployment, it should just work.

This workshop was created to help you to go from no knowledge of GitOps, to running a multi-tenant and efficient deployment procedure, with automatic and self-healing application deployments.

You need to have some knowledge about Kubernetes concepts, like how to deploy resources and write manifests, and foundational knowledge of Git for this workshop.

In this workshop, the attendees get to experience setting up multiple environments from a mono-repository, based on Kustomize and GitOps best practices. We will go through multi-tenant deployments, automation of container tags, and progressive delivery with Flagger.

This workshop was first held during KCD Oslo 2024, and received good feedback.

Valgets kvaler – bygg, kjøp, eller lei? en no

Plattformstrategi, det er enkelt det! Eller, kanskje ikke? Det finnes tross alt tonnevis med ferdige løsninger, åpen-kildekode prosjekter, og ikke minst meninger der ute.

Det er mange som mener at en plattform må bygges fra bunnen av, noen mener man må kjøpe så mye som mulig som tjenester, og noen vil bare møte opp til dekket bord. Det er fordeler og ulemper med alle disse, og når vi også har en bransje i konstant bevegelse – da blir det fort vanskelig å treffe blink.

Valget mellom å bygge, kjøpe eller leie er sjeldent enkelt, men det er en avgjørende strategisk beslutning. Det finnes ingen fasit, og det er ikke alltid man har kun en plattform å ta valget for.

Men hva er fordelene og ulempene med de forskjellige tilnærmingene, og i hvordan retning går trendene i disse dager? Er det egentlig så dyrt å bygge, og hvordan kan man unngå å kjøpe katta i sekken? Dette foredraget tar sikte på å belyse nettopp disse spørsmålene, og gi deg noen konkrete perspektiver for å navigere i valgene og unngå de vanligste fallgruvene.