Cncf :: SREKubeCraft | Nick Nikolakakis

~/posts/modelplane-fleet-inference 29 min · 5971 words

$ Modelplane - Fleet Inference Across Clusters, Demoed Without a Single GPU

// A hands-on tour of Modelplane, the Crossplane-based control plane for AI inference across a fleet of GPU clusters. Two regions on Kind with fake DRA GPUs, the platform/developer contract written in DRA's own vocabulary, fleet-level drain, and the release trap that costs you two APIs.

~/posts/agent-sandbox 19 min · 3912 words

$ Agent Sandbox - Kubernetes-Native Sandboxes for AI Agent Code Execution

// A hands-on tour of agent-sandbox, the Kubernetes SIG Apps project for isolated, stateful, singleton workloads built for AI agent runtimes. Includes a Go agent with sandboxed tools on Kind, warm pools claiming in 53ms, Flux GitOps, and the gotchas the docs don't mention.

~/posts/llm-d-distributed-inference 21 min · 4271 words

$ llm-d - Kubernetes-Native Distributed LLM Inference at Scale

// A hands-on tour of llm-d, the CNCF Sandbox framework for distributed LLM inference on Kubernetes - inference-aware routing, prefill/decode disaggregation, and KV-cache offload. Includes a GPU-free demo on Kind using the vLLM simulator, wired with Flux GitOps.

~/posts/kserve 20 min · 4183 words

$ KServe - Production ML Serving on Kubernetes, from sklearn to LLMs

// How to use KServe v0.18 in Serverless mode to serve both classical ML models and LLMs on Kubernetes - with scale-to-zero, OpenAI-compatible APIs, Flux GitOps, and a real demo on Apple Silicon.

~/posts/dapr 20 min · 4245 words

$ Dapr - Building a Safe Platform for Citizen Developer Apps on Kubernetes

// How to use Dapr, Flux, Harbor, and a shared Helm chart to safely host apps built by non-technical citizen developers using LLMs on Kubernetes.

~/posts/oauth2-proxy 21 min · 4283 words

$ OAuth2-Proxy - Securing MCP Servers on Kubernetes Before Hackers Find Them First

// MCP servers like DBHub expose databases, filesystems, and code execution over HTTP with zero authentication. Learn how to deploy OAuth2-Proxy on Kubernetes to add SSO, group-based access control, and session management to any MCP server without changing a single line of code.

~/posts/knative 15 min · 3111 words

$ Knative - The Platform Engineer's Guide to Serverless on Kubernetes

// A guide to Knative for SREs and Platform Engineers. Learn how Knative Serving, Eventing, and Functions bring serverless capabilities to any Kubernetes cluster with autoscaling, scale-to-zero, and event-driven architectures.

~/posts/k8s-multi-tenancy 21 min · 4444 words

$ Building a Multi-Tenancy Platform with Capsule and vCluster - Hard vs Soft Isolation

// Learn how to build secure, scalable multi-tenant Kubernetes platforms using Capsule for namespace-based isolation and vCluster for virtual cluster deployment. Compare architectures, explore production use cases, and implement cost-effective multi-tenancy patterns with practical examples and complete demos.

~/posts/kubernetes-autoscaling-beyond-hpa 24 min · 5026 words

$ Kubernetes Autoscaling Beyond HPA - Event-Driven KEDA and VPA Resource Optimization

// Master advanced Kubernetes scaling strategies with KEDA for event-driven autoscaling and VPA for resource optimization. Real production examples, cost savings, and complete implementation guide for SRE teams.

~/posts/litmuschaos 9 min · 1770 words

$ Litmus Chaos - Enterprise-Grade Chaos Engineering for Kubernetes

// Discover Litmus, a CNCF incubating project that provides a complete chaos engineering platform with GitOps integration, extensive experiment library, and enterprise features for building resilient cloud-native applications.

posts tagged #Cncf