Overview
APE XXL 2026 is the biggest Azure & AI community day in the Netherlands, offering a full day of talks, workshops, a keynote, dinner, and drinks. It brings together engineers building AI platforms to learn about AI agents, governed AI platforms, Bicep, Fabric, and security.
Details
- Call for talks
- OpenSubmit a talk
Speakers
19- Alina AstapovichPlatform Engineer at Storytel
- Bastiaan WassenaarCloud Infrastructure Architect
- Danny KrugeMission Critical Engineer at Schuberg Philis, Microsoft Data Platform MVP
- Dennis SchooneEnterprise Cloud Architect and MCT, founder of CloudMatters
- Erwin DerksenFounder of DerkIT and Microsoft Certified Trainer
- Gijs ReijnCloud Engineer at Rabobank
- Hans MunsFreelance .NET/Azure Engineer & Microsoft Certified Trainer @ Backnook
- Hidde de SmetAzure Architect at Xebia and Microsoft MVP
- Jesse WellenbergSoftware and Cloud Architect at Xebia
- Leo VisserSubject Matter Expert Azure and AI Foundry, Microsoft PowerShell MVP
- Maik van der GaagHead of Technology Microsoft at Databalance, Microsoft Azure MVP
- Marc DekeyserFounder of Crimson Owl Technologies
Show all 19 speakers
- Mark ScholmanFreelance Azure Consultant and Microsoft Azure MVP
- Mello BogaartsData & AI Engineer at Omring
- Niels KokTechnical Consultant, Microsoft MVP and MCT
- Patrick de KruijfAzure Architect and Global Landing Zone Solutions Lead at Xebia
- Rex de KoningAutomation Consultant and MCT, Microsoft Azure MVP
- Stefan DingemanseModern Workplace Consultant and Architect, Microsoft MVP
- Wyke NieuwenhuizenManager Microsoft Competence Center at Omring
Schedule
12 security checks every Microsoft 365 / Entra ID admin must do
In many organizations the obvious settings are never evaluated or checked — and that creates avoidable security exposure. A visual, #Bullet Point Free run-through of twelve checks every admin should do to make sure the basics are actually covered. Applies to Entra ID as well as Microsoft 365.
Why is your AI agent ignoring the instructions you just wrote?
copilot-instructions.md, AGENTS.md, SKILL.md, prompt files — they all tell the agent how to behave. So why does one get followed immediately while another disappears into the void? A practical mental model for how AI instruction files are actually loaded and applied — and how to spot guidance that is being ignored versus never loaded.
Azure Enterprise Policy as Code (EPAC)
Managing Azure Policy through the portal becomes a hassle once your tenant reaches a certain size — and with new AI services landing almost every month, good governance is key. A level-400 walk-through of Enterprise Policy as Code: desired-state strategy, managing policies, policy sets and assignments, and policy documentation as code.
Azure Networking Deep Dive & Platform Engineering in the AI Era
This workshop opens with a level 400 deep dive into Azure networking. Expect in-depth coverage of Azure network architectures, routing, hub-spoke, vWAN, private connectivity, DNS, security patterns, and troubleshooting at scale. Not an introduction — a session for people who genuinely want to understand and design Azure networks at enterprise level. In the second half we shift focus to the future: the intersection of AI and platform engineering. How do you build platforms that are not only stabl
How We Stopped Our AI Agents From Lying
A build-and-break story from production: 32 specialized AI agents assess a live Azure estate against the Well-Architected Framework — and the hardest part was never the prompt. Watch the system produce a plausible finding that is wrong, then catch its own hallucination live. Three reusable patterns for agent output you can actually trust.
Terraforming Fabric: From Zero to Workspace Hero
Every Microsoft Fabric implementation starts the same way: someone clicks "New Workspace". That works for one. But what about 50? 200? With DEV, ACC and PROD for each? How Fabric became a fully automated, enterprise-ready platform using Terraform — a workspace vending solution where one definition provisions everything a production-ready environment needs.
How I Escaped the Token Bill and Moved to Local AI
Most engineers default to cloud AI without questioning the cost, the privacy trade-offs, or the dependency it creates — and the token bill grows with usage. How one freelancer moved an AI-assisted development workflow from 100% cloud to 90% local: the hardware, the models that actually perform, the honest downsides, and where cloud still wins.
Break
How Memory Works, from the Human Brain to AI Agents
AI agent memory sounds complicated until you realize there is already a perfect example of it: your own brain. Each type and technique of agent memory matched to the part of the brain it mirrors — then shown at work in a real agentic system in production: how it decides what to remember, how it recalls the right thing at the right moment, and how it forgets the rest.
Dinner
From 10 to 10,000 Agents: Engineering a Governed AI Platform on Azure
Building one AI agent is easy; engineering a platform for thousands is another story. Which department is burning the most tokens? Does every agent in production follow your safety guidelines? A reference architecture for a centrally governed AI platform on Azure — token rate limiting, semantic caching and security through API Management and Microsoft Foundry, agent identity in Entra ID, deployed as a hub-and-spoke landing zone in code.
The Dawn of the Mesh: Zero Trust at scale
For years hub-and-spoke has been the familiar backbone of cloud network design — predictable, centralized, and increasingly limiting. As organizations accelerate and embrace Zero Trust, the old patterns show their cracks. Why the future belongs to mesh-based architectures: decentralized, identity-driven, policy-enforced, and built for the scale of modern cloud.
Microsoft Intune and AI: Build Your Own AI Assistant
Security Copilot looks impressive, but the price tag might not be. Build your own AI assistant for Microsoft Intune using Azure services and Microsoft Foundry — one that understands your Intune environment and gives actionable advice on configurations, conflicts, and compliance issues, without the enterprise price tag.
From healthcare data to AI in production: building with Microsoft Fabric, AI Foundry and LLMs at Omring
Healthcare has to do more with the same number of people. At Omring, AI is not an experiment but part of the digital strategy. A look under the hood of their AI platform: Microsoft Fabric, Lakehouse, Py Spark notebooks, Azure AI Foundry and LLMs combined into working solutions — plus the architecture choices and lessons from idea to production.
Clean and Controlled AI with API Management
Everyone uses AI to improve their organization, but without care you drain your budget, increase energy usage, or create risks you never intended. Do you keep a leash on your AI? Practical governance and sustainability for your AI landscape with Azure API Management and Azure AI Foundry: content filtering, token-aware throttling, and caching.
Backstage Bicep: Exploring What's Next Before It Ships
Every Bicep release ships experimental features you can try — but some innovations are not in an official release at all. They sit behind a feature flag, or quietly wait to be merged from a pull request. A guided tour of what has not landed yet: installing the nightly CLI, running the new syntax improvements, and the behind-the-scenes work that changes how you write Bicep. Not for production.
Drinks
12 security checks every Microsoft 365 / Entra ID admin must do
In many organizations the obvious settings are never evaluated or checked — and that creates avoidable security exposure. A visual, #Bullet Point Free run-through of twelve checks every admin should do to make sure the basics are actually covered. Applies to Entra ID as well as Microsoft 365.
Why is your AI agent ignoring the instructions you just wrote?
copilot-instructions.md, AGENTS.md, SKILL.md, prompt files — they all tell the agent how to behave. So why does one get followed immediately while another disappears into the void? A practical mental model for how AI instruction files are actually loaded and applied — and how to spot guidance that is being ignored versus never loaded.
Azure Enterprise Policy as Code (EPAC)
Managing Azure Policy through the portal becomes a hassle once your tenant reaches a certain size — and with new AI services landing almost every month, good governance is key. A level-400 walk-through of Enterprise Policy as Code: desired-state strategy, managing policies, policy sets and assignments, and policy documentation as code.
How I Escaped the Token Bill and Moved to Local AI
Most engineers default to cloud AI without questioning the cost, the privacy trade-offs, or the dependency it creates — and the token bill grows with usage. How one freelancer moved an AI-assisted development workflow from 100% cloud to 90% local: the hardware, the models that actually perform, the honest downsides, and where cloud still wins.
How We Stopped Our AI Agents From Lying
A build-and-break story from production: 32 specialized AI agents assess a live Azure estate against the Well-Architected Framework — and the hardest part was never the prompt. Watch the system produce a plausible finding that is wrong, then catch its own hallucination live. Three reusable patterns for agent output you can actually trust.
Terraforming Fabric: From Zero to Workspace Hero
Every Microsoft Fabric implementation starts the same way: someone clicks "New Workspace". That works for one. But what about 50? 200? With DEV, ACC and PROD for each? How Fabric became a fully automated, enterprise-ready platform using Terraform — a workspace vending solution where one definition provisions everything a production-ready environment needs.
How Memory Works, from the Human Brain to AI Agents
AI agent memory sounds complicated until you realize there is already a perfect example of it: your own brain. Each type and technique of agent memory matched to the part of the brain it mirrors — then shown at work in a real agentic system in production: how it decides what to remember, how it recalls the right thing at the right moment, and how it forgets the rest.
The Dawn of the Mesh: Zero Trust at scale
For years hub-and-spoke has been the familiar backbone of cloud network design — predictable, centralized, and increasingly limiting. As organizations accelerate and embrace Zero Trust, the old patterns show their cracks. Why the future belongs to mesh-based architectures: decentralized, identity-driven, policy-enforced, and built for the scale of modern cloud.
From 10 to 10,000 Agents: Engineering a Governed AI Platform on Azure
Building one AI agent is easy; engineering a platform for thousands is another story. Which department is burning the most tokens? Does every agent in production follow your safety guidelines? A reference architecture for a centrally governed AI platform on Azure — token rate limiting, semantic caching and security through API Management and Microsoft Foundry, agent identity in Entra ID, deployed as a hub-and-spoke landing zone in code.
Microsoft Intune and AI: Build Your Own AI Assistant
Security Copilot looks impressive, but the price tag might not be. Build your own AI assistant for Microsoft Intune using Azure services and Microsoft Foundry — one that understands your Intune environment and gives actionable advice on configurations, conflicts, and compliance issues, without the enterprise price tag.
Backstage Bicep: Exploring What's Next Before It Ships
Every Bicep release ships experimental features you can try — but some innovations are not in an official release at all. They sit behind a feature flag, or quietly wait to be merged from a pull request. A guided tour of what has not landed yet: installing the nightly CLI, running the new syntax improvements, and the behind-the-scenes work that changes how you write Bicep. Not for production.
Clean and Controlled AI with API Management
Everyone uses AI to improve their organization, but without care you drain your budget, increase energy usage, or create risks you never intended. Do you keep a leash on your AI? Practical governance and sustainability for your AI landscape with Azure API Management and Azure AI Foundry: content filtering, token-aware throttling, and caching.
From healthcare data to AI in production: building with Microsoft Fabric, AI Foundry and LLMs at Omring
Healthcare has to do more with the same number of people. At Omring, AI is not an experiment but part of the digital strategy. A look under the hood of their AI platform: Microsoft Fabric, Lakehouse, Py Spark notebooks, Azure AI Foundry and LLMs combined into working solutions — plus the architecture choices and lessons from idea to production.
12 security checks every Microsoft 365 / Entra ID admin must do
In many organizations the obvious settings are never evaluated or checked — and that creates avoidable security exposure. A visual, #Bullet Point Free run-through of twelve checks every admin should do to make sure the basics are actually covered. Applies to Entra ID as well as Microsoft 365.