Blog
Beyond the Model: The Hardware Fundamentals That Define Your AI Strategy
How does model size translate into real hardware requirements?
In this post, we break down the fundamentals every tech professional should know about LLM sizes (overview and intended use), memory demand (how to estimate it quickly and reliably), hardware choices and the VRAM bottleneck during inference.
Dagger: CI/CD as Code and Agentic AI enabler
CI/CD pipelines are supposed to help developers ship better code, and faster. In practice, they quite often do the opposite. Developers still need to write scripts to build and test applications locally. Environment configurations bloat pipelines with opaque and hard-to-reuse YAML code. And as workflows expand beyond CI/CD to integrate agentic AI, traditional tools start to show their limits. Dagger was created to address exactly these problems.
One Prometheus to Rule Them All: Multi-Tenancy Kubernetes with Centralized Monitoring and vCluster Private Nodes
Discover how platform teams can implement centralized metrics for multi-tenant Kubernetes using vCluster. This article walks through observability patterns for both regular vClusters and private-node vClusters, showing how a centralized Prometheus and Grafana stack can serve many isolated tenant clusters, laying the foundation for scalable, production-ready multi-tenant observability.
Isolated GPU Nodes on Demand: Implementing vCluster Auto Nodes for AI Training on GKE
Learn how to provision isolated GPU nodes on demand for multi-tenant AI training on GKE. This tutorial implements vCluster Auto Nodes with Private Nodes, giving each tenant dedicated Compute Engine VMs that spin up automatically and terminate when workloads complete. Cost-efficient GPU isolation without managing separate clusters.
MCP-Server auf Kubernetes mit kmcp und AgentGateway angehen
Verstehen Sie das Model Context Protocol, seine Verwendung, Funktionalität, wie man es sichert und warum es bald zu einem der grundlegenden Bausteine in einer KI-Agent-Architektur werden wird. Entdecken Sie verschiedene Möglichkeiten und Beispiele für die Einrichtung von MCP-Servern in Kubernetes mit AgentGateway, kgateway und kmcp. Erfahren Sie, wie Sie verschiedene MCP-Server unternehmensfähig machen, indem Sie sie mit Unternehmensfunktionen wie Authentifizierung, Autorisierung und Skalierbarkeit ausstatten.
Taming the Multi-Cloud Chaos: Meet Scaleway Kosmos
Feeling locked into a single cloud provider? Wishing you could seamlessly use resources from different clouds without the operational nightmare?
That's exactly what Scaleway Kosmos was built to solve.
