Who's Really in Control? AI Agents Meet Software Engineering

We're excited to announce our next meetup May 28th, 2026! As usual, we'll take care of food, drinks and awesome sessions, you just need to bring you ;-)

This meetup will feature three talks aimed at (Java) developers, and they all revolve around the question: what does software engineering look like in the age of AI?

• Bert Jan Schrijver kicks off with a talk on who's really in control when AI agents start writing and changing your code
• Tom Wigleven closes the evening by sharing how to accelerate your entire software delivery pipeline with AI agents, without sacrificing quality
• Mitchell Tieleman & Dean Falix from ODIN Labs then show how to orchestrate long AI workflows without hallucinations, with a live demo

All sessions will be presented in English; RSVP now for an evening of learning and fun! The meetup will be at the OpenValue office in Utrecht (directions: https://route.openvalue.eu), enough free parking spaces available.

Agenda

• 17:30 Walk in
• 18:00 Food
• 18:30 Talk 1 (see below)
• 19:00 Break
• 19:15 Talk 2 (see below)
• 19:45 Break
• 20:00 Talk 3 (see below)
• 21:00 Drinks

First talk: Software engineering in the age of AI agents: who's in control?
An AI deletes a production database… and insists it didn't. Welcome to modern software engineering. Generative AI is becoming a powerful collaborator: writing code, making changes, taking action. But it doesn't always do what we think it does. In this keynote, we'll explore what's really changing: from writing code to verifying it, from building to guiding, and from trusting systems to critically reviewing them. I'll share personal stories, industry insights, successes and failures, and how to responsibly use rapidly evolving AI tools. Because even if you let AI do the work, you're still accountable for the outcome.

About Bert Jan:
Bert Jan is CTO at OpenValue and focuses on Java, AI, software architecture, Continuous Delivery and DevOps. Bert Jan is a Java Champion, JavaOne Rock Star speaker, Duke's Choice Award winner and leads NLJUG, the Dutch Java User Group. He loves to share his experience by speaking at conferences, writing for the Dutch Java magazine and helping out Devoxx4Kids with teaching kids how to code.

Second talk: How to Accelerate Software Delivery Without Sacrificing Quality
"AI writes code now." Sometimes it's good. Sometimes it confidently writes something broken. The gap between those two outcomes isn't the model. It's how you use it.
We can automate and orchestrate almost everything in the software development workflow with the right agents and skills. How can we confidently increase the quality of the output? Structured specs, agent teams that design, implement and review each other's work, guardrails, extensive e2e testing, and a self-improvement loop that makes every run better than the last. The result is close to a full development team at your disposal, but remember you're accountable for what ships! You set the direction and control the quality.

About Tom:
Tom Wigleven has spent twenty years writing software. These days he writes specs and lets agents do the typing. At OpenValue he's involved with AI accelerated development. He talks about the workflows, the guardrails, the quality, the responsibilities and what it means for how teams ship software.

Third talk: Own Your AI, Not Theirs: Orchestrate (Long) Workflows Without Hallucinations
OpenValue asked: "What happens when AI deletes a production database and insists it didn't?"

Odin Labs answers: Build AI that actually knows what it's doing.

In this session, we demonstrate what makes Odin Labs unique: orchestration of long, complex workflows powered by BrainDB - organizational memory that eliminates hallucination at the source. You'll see a live demo of an AI agent executing multi-step work orders with full context retention across the entire workflow. No context drift. No guessing. No black boxes.

Core message: When AI has the right context, it doesn't need to guess.

About Mitchell & Dean:
Mitchell Tieleman (Co-Founder & CTO) and Dean Falix (Co-Founder & CEO) run ODIN Labs, a Netherlands-based AI company building on-premise AI agent infrastructure for regulated industries like healthcare, legal, finance and government. Their core philosophy: Own Your AI. Not Theirs. Everything they build is GDPR-native, free of vendor lock-in, and governance-first by design.