Not just another AI talk: MCP - a standardized protocol for AI

Model Context Protocol (MCP) defines itself as "a USB-C port for AI applications". It standardizes the way you can connect your devices to various peripherals and accessories. A standardized way you connect AI models to different data sources and tools.

This presentation will show you, in a practical demo, how you can develop clients and servers to make use of agentic AI as their user interface. Your MCP servers will make it possible for other developers to directly plug into your services. But be aware, that means it's easier for your customers to switch to your competitors as well!

We will show you best practices for securing your data within your own infrastructure, and how to control your data flow. May the demo monster be with us.

We will also address some common pitfalls, and what you need to consider before diving head first into making your very first MCP server, exposing all your company's deepest secrets.

Challenge yourself to make your services and data available in a whole new way, to be the cool kid and follow the latest trends without going face first into a brick wall.

Mainly live demo. Target audience is everyone.

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

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!

Managing the Chaos: Not a Superpower. Not a Curse. Just My Brain

Join me for a fun and honest talk about living with ADHD. Something I didn't fully understand about myself until later in life. I'll share my personal journey from feeling like a useless "scatterbrain", to learning how to work with my brain instead of against it.

This isn't a medical lecture, but rather a real-life story with humor, practical tips and a few nerdy metaphors (yes, we'll talk about RPG and mana use and regeneration, my "mana theory"). We will explore how I experience the world, with my flavour of neurospice, what makes our brains different and how workplaces can become more inclusive and creative by embracing those differences.

Whether you're neurodivergent yourself (or neurospicy as I like to call it), wotk with someone who is, or just want to understand people better - maybe this is a good talk for you.

Go with the Flow: Stop the Leaks Before Your Memory's a Waterfall!

You thought you didn't have tio think about memory management and memory leaks when you were done with C? Think again!

Join me starting the dive by looking at how golang allocates and deallocates memory, and the workings of the garbage collector. Learn how the stack and heap works. And why does my substring allocate so much memory?

We'll dive into some pitfalls we all are guilty of, and some that are real edge cases. We're looking at code, and coding patterns. What can cause your application, website or k8s operator to consume increasingly amounts of memory? And how to avoid it without crashing out in a OOM (Out of memory) error?

And can I actually answer all of my own questions? Maybe more? Will you even know without joining in?

MCP Security: Keep Your AI Agents from Spilling the Tea

Model Context Protocol (MCP) is often referred to as "the USB-C port for AI applications". Just like REST is our go-to standard for application-to-application integrations, MCP is really running up as a more or less de facto standard for agent-to-application integrations. But how does one secure these endpoints? Just like your REST-endpoints?

We will take you through a mixture of practice (yes, we will do live demo) and theory of authentication in MCP Servers. We will look into the "how", "why" and "when".

We will also address some common pitfalls, issues and limitations, like "consent fatigue" (what is that? Join, and we'll tell you). Can you just re-use your existing RBAC thought, or does the nature of agents and MCP endpoint structure force you to start thinking different?

Make your services and data available in a whole new way, hopefully without exposing all your company's deepest secrets. Let us give you some concrete actions and bullet points for you to bring back and get started!