Thank you to each of you for making Cloud Native Zürich a resounding success. Our sincere thanks also go to our sponsors. Your support played a crucial role in the success of this event.
Even though vector databases are critical infrastructure for AI applications, most teams only consider backup strategies after disaster strikes. Whether it's a corrupted index, accidental deletion, or a ransomware attack, losing vector embeddings can cripple RAG systems and leave LLMs hallucinating or returning empty results. In this session, we'll explore what actually works (and what doesn't) when protecting high-dimensional data, walk through Kubernetes-native strategies for backing up and restoring vector databases, and show a real-world recovery scenario that proves resilience doesn't have to come at the cost of performance.
Build GitLab CI runners at Mobiliar run on an AKS Kubernetes cluster (with autoscaling). However, the jobs vary significantly in terms of resource requirements (CPU, memory). As infrastructure operators, we want to ensure that every job gets the resources it needs to run quickly and successfully, while also making sure the infrastructure isn't oversized. Developers often don't know exactly how many resources their jobs require. Additionally, GitLab's built-in tools make it difficult to manage appropriate resource requests across all Mobiliar projects (and there are many). To address this, we've developed a solution using a Kubernetes Mutating Webhook combined with Prometheus lookups and caching. Based on historical job data, it automatically sets suitable resource requests when a pod is created. We'd be happy to share more context and details about this approach in a talk.
Platform engineering has established itself as a reliable approach for the cloud native tech generation. But what really makes a standout platform? Is it the technology? Is it a product focus? And how do we measure “organizational success” from a platform point of view? Rather than theorize, let's look at how large, forward-thinking organizations like Adidas, Spotify, ING or SAP are leveraging platform engineering to accelerate and simplify the delivery of software at scale. Turns out a good platform approach not only provides helpful, easy to use services, but can actually unlock organizational bottlenecks and improve team dynamics and interactions (aka “the culture”). Some proven patterns include organizing teams along value streams inside coherent platform groupings, setting expectations for platform interactions (beyond the APIs!), focusing on business and flow metrics, intentionally addressing consumer/provider mismatches, a platform manifesto for shared understanding, and more! Join this talk to unlock your platform engineering success!
Thanks to our sponsors for CNZ 2025
We would like to extend a special thank you to our sponsors for their support and commitment to the community. We couldn't do it without them! If you're interested in becoming a next year sponsor, please contact hello@cloudnativezurich.ch.
Gold
