26 - 28 OCT 2026 Utrecht Netherlands

AI & Agents

Ship AI your users can trust

Would you trust your AI agent with a real user?

AI becomes useful when it can operate safely in the systems people already depend on. This journey follows the engineering decisions that turn an agent from a promising proof of concept into a product your team can operate with confidence.

This journey is for

Software engineers building AI features, agents and intelligent applications.

What you will be able to do

Make better choices about identity, reliability, safety and evaluation before your AI system reaches real users.

Recommended sessions

Follow the route

Start with the sessions below, then use the full agenda to make the route your own.

  1. Beat the Clock: Building a Production-Ready Agentic App in 60 Minutes

    Jeff Prosise

    Start with a practical build that shows what production-ready agentic application work actually involves.

  2. POC Prison: Why agentic systems never escape the lab and how to fix that in 90 days

    Luise Freese

    Learn why promising agent prototypes stall and what gives them a credible route to real value.

  3. Identity propagation for AI Agents: From user to tool

    Arne De Proft, Laura Verghote

    Bring user identity and permissions with the agent all the way to the tools it can use.

  4. Reliable Agentic Systems need Durable Execution

    Marc Duiker

    Explore the reliability choices an agentic system needs when work cannot simply disappear or be retried by a person.

  5. Applying the OWASP Top 10 for Agentic Applications to Your AI Agents

    Will Velida

    Turn AI safety into concrete engineering checks for permissions, context and abuse paths.

  6. Continuous Evaluation & Monitoring for AI Applications

    Soham Dasgupta

    Close the loop with ongoing evaluation and monitoring once the system meets real usage.

Optional stops

  1. Building a Hybrid AI Platform for Agent-Heavy Engineering

    Willem Meints

    An optional platform perspective on running agent-heavy engineering work beyond a single prototype.

  2. Use Agents to Build Agents with Aspire + Microsoft Agent Framework

    Maddy Montaquila, Tommaso Stocchi

    A useful detour for teams connecting their agent work to a practical .NET application foundation.

Take this back to your team

Make the next conversation more useful.

  1. A practical definition of production-ready for your next AI initiative.
  2. Sharper questions about what an agent is allowed to access and do.
  3. A starting point for testing and monitoring AI behaviour after release.
View agenda Get your ticket