The Human Side of Code & The Silent Failures of AI

We are thrilled to welcome you to our 10th edition of the AI Native Netherlands, hosted at the Yuma headquarters in Amsterdam.

This edition is capped at 120 attendees. Yuma's office has limited space, so once the 120 seats are filled, registration closes and additional signups move onto the waitlist.

A massive thank you to our host, Yuma, for their generosity in providing the venue, food, and drinks for this evening and for supporting the Dutch AI engineering community.

We'll cover:

• What humans still bring to software building when AI is writing the code
• Tackling technical and social complexity inside distributed, fast-moving systems with legacy at the core
• How to evaluate AI agents in production and catch silent regressions in under an hour
• What "AI Native infrastructure" looks like in the wild: prompt changes, harness updates, RAG documentation, model swaps
• A smaller, capped edition with the speaker Q&A returning at the end of the evening

Speaker 1: Nico Krijnen (Yuma)
An explorer at heart, Nico loves to navigate the ever changing technology landscape and find practical solutions to complex problems. For a big chunk of his career, he disrupted the DAM space by building an innovative file management solution called Elvis, which is being used by media organizations all over the world. Besides leading and supporting teams and giving direction, he likes to roll up his sleeves and experience first-hand how (and whether) the latest technologies work in the real world. He is never shy of making unconventional choices to achieve radical results.

Talk: The human side of coding
Now that AI's are writing our code, what is the value that we humans still bring to building software systems?

Our tech stacks are exploding in complexity, systems are more and more distributed and fast-paced development leads to an ecosystem full of fragile legacy systems that provide the backbone of business capabilities. How do we tackle that technical and social complexity? Do we take the usual route and solve it by adding more tech or AI? Or is there another way?

Even in the 80's, people realized that the theory of a system can't be transferred through code and documentation alone. It requires developers to work closely together. We need to encourage human collaboration across disciplines, backgrounds and personalities (including our new AI buddies). Learn how you can use your human power skills to better understand the problems you're facing and build systems that last!

Speaker 2: Floris Weers (Biscuit)
Floris is an ML engineer based in Amsterdam. He was responsible for continued-pretraining (or mid-training) at Apple's Foundation Model team, contributing directly to Apple Intelligence. Before that, he finished his master's degree at the University of Twente while building games on the side.

He is a named author on the Apple Intelligence Foundation Language Models tech report, which describes the ~3 billion parameter on-device model and the large server-based model powering Apple Intelligence features. His research work at Apple also spanned multimodal models, mobile vision transformers, and distillation scaling laws.

Talk: The silent failures of AI: how to evaluate?
AI doesn't crash when it fails. It quietly does half the job. We'll go over a systematic way to evaluate agents in real product settings, allowing us to catch regressions and decide within an hour whether to use a newly released model or stick with the previous setup. A side-effect of the evaluation framework is that no engineer is scared of touching system prompts, making deeper harness changes or adding documentation to the RAG systems.

Agenda:
18:00 — Arrival, food & drinks
18:30 — Talk #1 | Nico Krijnen
19:15 — Talk #2 | Floris Weers
20:15 — Speaker Q&A with Nico and Floris
20:45 — Networking & more drinks
21:00 — Wrapping up

What to bring: Just curiosity and questions. If you've been working through agent evaluation, RAG system updates, or human-AI collaboration in your engineering team, bring those questions for the speaker Q&A.

Who is this for: Platform engineers, AI/ML engineers, SREs, architects, and engineering leaders focused on building and validating reliable AI systems in production.

Where to find us: Yuma (4th floor), Business Centre Pedro de Medina, Pedro de Medinalaan 1, Amsterdam. 15 minutes from Amsterdam Central using tram 26. Paid parking at Parkbee Sluishuis across the street. Maps link