Skip to content
MeetupIn personPast event

From Prompts to Pre-Training: Quality, Scale, and the Road to AI Native

Event artwork for From Prompts to Pre-Training: Quality, Scale, and the Road to AI Native

This event has ended

This listing is kept as an archive. Explore upcoming events to find your next one.

See upcoming events in Amsterdam
About this event

What to expect

Building AI systems that actually work in production requires more than prompt engineering—it demands rigorous quality frameworks, scalable infrastructure, and architectural patterns that can handle real-world complexity. This Amsterdam meetup brings together two expert practitioners to share battle-tested approaches for shipping reliable LLM-based systems at scale.

What You'll Learn

From Prompts to Production: Engineering Quality in LLM-Based Systems by Ronny Roeller (NEXT AI) tackles the fundamental challenge every AI builder faces: how do you ensure consistent quality when outputs are probabilistic and data-dependent? Ronny, CTO of NEXT AI—a customer insights platform serving clients like Deel, Bosch, and Rituals—will walk through concrete failure modes encountered in production, including silent quality degradation that emerges over time. You'll learn about multi-layer evaluators, human-in-the-loop safeguards, and architectural patterns for measuring, monitoring, and continuously improving LLM system quality beyond basic prompt tuning.

Towards The Adyen Foundational Model by Hanna van der Vlis shifts focus to foundational models at unprecedented scale. Adyen processes over 40 billion transactions annually, and Hanna is building AI systems that must deliver sub-30ms latency in production. She'll outline how Adyen is moving beyond traditional point-in-time feature engineering by treating payment sequences as a language problem, pre-training a unified backbone on up to 51.2 trillion tokens. Expect deep technical insights into overcoming compute bottlenecks, fine-tuning for downstream tasks like fraud detection, and meeting strict production SLAs at massive scale.

Interactive Q&A Panel

After both talks, submit your technical questions online and watch as speakers address them live during a structured panel discussion. This format ensures your specific challenges around LLM workflows, training infrastructure, or production reliability get direct attention from practitioners who've solved similar problems.

Who Should Attend

This event is designed for platform engineers, AI/ML engineers, SREs, architects, and engineering leaders focused on building and validating reliable AI systems in production. Whether you're wrestling with LLM evaluation strategies, scaling training infrastructure, or maintaining quality in agentic pipelines, you'll find relevant technical depth and practical patterns

Running an AI event in Amsterdam?

List it here, free.

Meetups, workshops and conferences go in front of thousands of builders after a quick review.

List your event