AI Spring Jam - speakers announced
Session Details
đź“… When? Monday, April 20, 2026
⏰ What time? 17:00 – 21:00
đź“Ť Where? Xebia HQ, Hessenstraatje 1F, Antwerp, Belgium
▶️ Sign up here
Round 1
1. The Lazy Architect's Guide to AI Agents
Speaker: Pieter Neys
AI agents don't have to be complex. In this session, I'll show you how to build surprisingly capable AI workflows with n8n and OpenRouter: pipelines that trigger themselves, think in steps, and actually get things done.
No theory, no slide marathons. Just live building, and a fresh perspective on why "lazy" is sometimes the smartest architectural choice you can make.
2. Building the AI Agent Harness
Speaker: Peter Eysermans
Most teams working with AI agents focus on prompting. The real skill is building the harness: the system of skills, hooks, and conventions that makes agents reliable without micromanaging every step.
In this session, I'll show you what I believe will be the next step: an agent harness that builds features from the feature passport up until the merged pull request. The system writes feature passports, implements them, runs QA, creates and reviews PRs, and resolves comments on PRs—all while keeping the human in the loop at every step.
Round 2
Multi Agent, Multi Model: A Story About Agentic Storytelling
Speaker: Frederiek Vandepitte
What happens when multiple AI agents collaborate to tell a story? In this session, we'll explore how agent orchestration across multiple foundation models can be harnessed for dynamic, interactive storytelling. Using Microsoft Agent Framework and deploying across multiple models hosted on Azure, we'll demonstrate how agentic AI systems can be designed to coordinate creativity, memory, and interaction.
The session will cover:
- The architecture of a multi-agent, multi-model setup
- The design principles behind agent roles in narrative generation
- Practical examples of collaborative storytelling in action
Whether you're a developer, designer, or storyteller, you'll leave with a clear understanding of how agentic systems can open new creative frontiers.
Round 3
1. POC Prison: Why agentic systems never escape the lab and how to fix that in 90 days
Speaker: Luise Freese
Most organizations experimenting with agentic AI aren’t blocked by models, frameworks, or orchestration. They’re blocked by something far more basic: the everyday realities of how enterprises actually work.
Agents look brilliant in controlled demos, but the moment you try to plug them into real systems (legacy data, governance, identity, compliance, and unclear ownership) they collapse into the same pile of abandoned POCs as everything else.
My talk cuts through the hype and gets straight to the uncomfortable truth:
- Why agentic systems end up as flashy prototypes instead of production tools
- How Excel-based “data estates” quietly choke autonomy before it even starts
- Why most so-called “AI use cases” are still rule-based automation wearing an AI sticker
- How to build the minimal delivery backbone needed for any intelligent agent to run safely in a real enterprise
This isn’t yet another vision talk, but the reality check most teams never get: the engineering and organizational work required to move agents from the lab into the world where the constraints are real and the stakes are even higher.
Care about getting agentic systems running rather than demoing? This session gives you the hard truths and the practical steps to finally make that possible.
2. Context Engineering 101
Speaker: Jan De Dobbeleer
You've heard of prompt engineering. You may already be good at it. Prompt engineering matters, but it is only one part of a larger system. Each time you talk to an AI, a selection process happens. Every token competes for attention, yet not everything carries the same weight.
The gap between frustrating AI output and useful AI output often depends on one factor: what the model has available at the exact moment it responds. Model size, temperature, and prompt wording help, but this isn't the full story. In this session, we'll cover why a larger context window does not guarantee better results, how repository structure affects output quality, and which habits lead to consistently strong responses.
Whether you're using GitHub Copilot, building agents, or trying to avoid repeating context every few minutes, this session gives you a practical mental model you can apply immediately.