Skip to content
MeetupIn personPast event

AI Security Night Amsterdam: Securing LLM Applications & Multi-Tenant AI Platforms

Tools & platforms

snyk
Event artwork for AI Security Night Amsterdam: Securing LLM Applications & Multi-Tenant AI Platforms

This event has ended

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

See upcoming events in Amsterdam

Supporting this event

Sponsors

Snyk
About this event

What to expect

Security challenges in AI applications are rapidly becoming critical as more teams move from prototypes to production. This technical evening in Amsterdam brings together security engineers, platform architects, and developers building AI-powered systems to explore practical approaches to securing LLM-based applications and multi-tenant AI platforms.

The evening kicks off with Brian Vermeer, Staff Developer Advocate for Snyk and Java Champion, presenting "Breaching LLM-Powered Applications: Overcoming Security and Privacy Challenges." This session dives into the unique security vulnerabilities that emerge when integrating LLMs into applications—from prompt injection attacks and API key leakage to data misuse and privacy violations. Brian will demonstrate how general security flaws can be exploited to manipulate LLM behavior and provide actionable strategies for building compliant, secure AI applications. With over a decade of software engineering experience and a background as a military reserve for the Royal Netherlands Air Force, Brian brings both technical depth and real-world security perspective to this critical topic.

The second talk, delivered by Luca Berton, a cloud-native and AI platform architect based in Amsterdam, tackles "Securing Multi-Tenant AI Platforms: Practical Lessons from OpenShift AI, GPUs, and MLOps." As organizations scale AI initiatives, the challenge shifts from individual model security to platform-level concerns. How do you safely enable multiple teams to train, experiment, and deploy models on shared infrastructure? Luca will share engineering-oriented lessons from building multi-tenant AI/ML environments on Kubernetes and OpenShift AI, covering workload isolation, GPU partitioning strategies, namespace design, access control patterns, supply-chain security, and governance for regulated environments. This session stays vendor-neutral and focuses on what platform teams need to know to enable AI adoption without compromising security.

What You'll Learn

  • Common security vulnerabilities in LLM-powered applications and how to mitigate them
  • Strategies for compliance and data privacy in AI systems
  • Practical workload isolation techniques for multi-tenant AI platforms
  • GPU resource management and partitioning for secure shared environments
  • Access control and namespace design patterns for Kubernetes-based AI workloads
  • Supply-chain security considerations for MLOps pipelines

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