Blog
Running Multi-Tenant GPU Workloads on Scaleway with vCluster
Namespace-per-team is the default way to share a GPU cluster. It's also the wrong one. This is what true multi-tenancy looks like in Terraform.
The AI Gateway Landscape: agentgateway, LiteLLM, Kong and Envoy AI Gateway
Throughout out recent events we have seen that AI Gateways have become the single most important entry point for companies. Because every team building with LLMs eventually hits the same issue: it starts with a single <Your Favorite AI Provider> key for one service and eighteen months later you have a dozen services, five model providers, a pile of MCP tool servers and no single place to answer "who's calling what, at what cost, under what policy."
Multi-Tenant Observability with vCluster: Centralized Metrics and Logs Using Prometheus, Loki, and Grafana
Discover how platform teams can implement centralized metrics and logging for multi-tenant Kubernetes using vCluster. This article walks through such an architecture for private-node vClusters, showing how a centralized observability stack can serve many isolated tenant clusters, laying the foundation for scalable, production-ready multi-tenant observability.
Stop Burning Your LLM Budget: Cost-Efficient LLMOps on Kubernetes
As LLM initiatives mature from pilot to production, infrastructure costs frequently scale faster than the value they deliver. The culprit is often operational inefficiency: idle GPUs, uncontrolled storage growth, cold start latency, and unplanned network egress. We examine the principal cost drivers in LLMOps on Kubernetes and provide actionable best practices across observability, GPU efficiency, throughput tuning, storage governance, and network topology.
Beyond the Model: The Hardware Decisions That Define Your AI Strategy
Unlocking the black box: How does model size translate into real hardware requirements?
In this post we break down the fundamentals every tech professional should know:
► LLM sizes: overview of LLMs and their intended use ► Memory demand: how to estimate it quickly and reliably ► Hardware choices: CPUs vs GPUs vs TPUs ► The VRAM bottleneck: why memory is the central constraint for inference performance
This is your guide to the engine room of Generative AI.
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.
