Blog

Understanding AI infrastructure and Kubernetes from first principles.

Deep-dive articles on Kubernetes, GPU infrastructure, and AI platform engineering — written by a practitioner.

AllAI InfrastructureGPUKubernetes
Why Kubeflow Trainer Exists: The Journey from Single GPU Training to Distributed AI on Kubernetes
KubernetesgpuJul 21, 2026

Why Kubeflow Trainer Exists: The Journey from Single GPU Training to Distributed AI on Kubernetes

Understand the evolution from single GPU training to PyTorch DDP and discover why Kubernetes needs Kubeflow Trainer to orchestrate distributed AI workloads.

IU
Isreal Urephu
Read →
How One Expired Certificate Quietly Stopped an Entire Kubernetes Control Plane
kubernetesJul 10, 2026

How One Expired Certificate Quietly Stopped an Entire Kubernetes Control Plane

A real production incident that explains how one expired API server certificate silently stopped Kubernetes reconciliation and brought an entire control plane to a halt.

IU
Isreal Urephu
Read →
Scaling LLM Inference on Kubernetes: Why Your Pods Take 30 Minutes to Start (And How to Fix It)
KubernetesGPUJul 7, 2026

Scaling LLM Inference on Kubernetes: Why Your Pods Take 30 Minutes to Start (And How to Fix It)

# Scaling LLM Inference on Kubernetes: Why Your Pods Take 30 Minutes to Start (And How to Fix It)

IU
Isreal Urephu
Read →
Understanding How Kubernetes Supports GPUs
KubernetesAI infrastructureJul 5, 2026

Understanding How Kubernetes Supports GPUs

Learn how Kubernetes discovers, schedules and exposes GPUs to containers, and why components like the Device Plugin and GPU Operator are essential for running AI workloads.

IU
Isreal Urephu
Read →