Notes from running Kubernetes fleets.
Practical writing on monitoring, cost, and troubleshooting — aimed at teams who run production Kubernetes without a dedicated SRE function.
Agentic AI Needs Policy-as-Code Before It Needs More Autonomy
As AI agents move from recommending fixes to executing them, Policy-as-Code frameworks like OPA are becoming the governance layer deciding what they may do.
Why VantraOps Sends Root Cause Analysis to Your AI Provider, Not Ours
AI root cause analysis usually means sending cluster data to a shared model. Why VantraOps routes it through your own AI provider, and what that protects.
AI SRE Agents: What Datadog, Dynatrace, and AWS Actually Shipped
Every major observability vendor shipped an AI SRE agent within months of each other in 2026. What changed, the early results, and the governance gap left open.
Label-Based Cost Allocation: The Fastest FinOps Win Available
Teams using label-based Kubernetes cost allocation cut unallocated cloud spend by roughly 23% in six months. Here's how the model works and where it breaks.
EKS vs AKS in 2026: What Kubernetes Actually Costs on Each
AWS EKS charges for the control plane, Azure AKS doesn't. Here's the real 2026 cost comparison — and why utilization matters more than either cloud's price.
The Average Kubernetes Cluster Runs at 8% CPU Utilization
New 2026 data shows average Kubernetes CPU utilization at 8%, with over-provisioning up from 40% to 69% year over year. Here's why it's getting worse.
In-Place Pod Resize Is GA: What It Actually Changes for Rightsizing
Kubernetes 1.35 made in-place pod resize stable. Here's why the old restart tax was the real blocker on rightsizing, and what's different now that it's gone.
Running LLM Inference on Kubernetes: A Small Team's Guide
Self-hosting LLM inference on Kubernetes without a dedicated ML infrastructure team — what actually matters for GPU scheduling, autoscaling, and cost control.
GPU Rightsizing: The Biggest Kubernetes Cost Lever Nobody's Using
GPU rightsizing is considered the largest underexploited Kubernetes cost lever in 2026. Here's how to find wasted GPU spend and what to actually do about it.
The 2am Page: Incident Response Without a Platform Team
No dedicated platform or reliability team? Here's a realistic incident response process for a small team, from the first page to a lightweight postmortem.
AIOps for Kubernetes, Realistically: What It Actually Automates
AIOps promises full automation but mostly delivers something narrower: automated correlation, triage, and context-gathering. Here's the honest version.
The FinOps Checklist for Teams That Don't Have a FinOps Team
Teams without a FinOps program waste an estimated 32-40% of cloud spend. A practical checklist covering visibility, rightsizing, cleanup, and governance.
Avoiding Vendor Lock-In With Kubernetes: A Realistic Take for Small Teams
Vendor lock-in inflates cloud costs 20-30% via proprietary APIs and egress fees. A realistic, low-effort way for small teams to stay portable.
Securing Your Helm Chart and Container Image Supply Chain
Supply chain attacks on Kubernetes intensified in 2026 — compromised images, Helm charts, and operator packages. Here's how to defend against them.
Kubecost Alternatives for Small Teams in 2026
Kubecost isn't a bad tool — it's built for a different scale. Here's how OpenCost, CAST AI, and bundled platforms compare for a small Kubernetes team.
GPU Scheduling on Kubernetes in 2026: DRA, Bin-Packing, and Avoiding Idle Spend
Dynamic Resource Allocation reached GA in Kubernetes 1.34. Here's what changed for GPU scheduling, and how small teams can avoid paying for idle accelerators.
Hiring Your First SRE vs. Buying Tooling: A Framework for Seed-Stage Teams
A competent SRE costs $150k-$200k a year. A practical framework for deciding when that hire makes sense versus when better tooling solves it.
Multi-Cloud Kubernetes Without the Multi-Cloud Headache
Most small teams don't need full multi-cloud Kubernetes. What's actually worth adopting from the playbook without doubling your operational surface.
What Runtime Security Actually Catches in Kubernetes
Static scanning checks images before they run. Runtime security watches what containers actually do. What eBPF tools like Falco catch that scanning misses.
OOMKilled in Kubernetes: Finding the Real Root Cause (Not Just Raising the Limit)
Raising the memory limit stops the paging, not the problem. A practical guide to diagnosing why a pod actually got OOMKilled.
PodDisruptionBudgets Explained: Why Your Deploys Keep Taking Down Prod
Missing PodDisruptionBudgets are a common cause of Kubernetes downtime during routine node drains and upgrades. Here's how to set them correctly.
The Four Golden Signals, Explained for Teams Without a Platform Team
Latency, traffic, errors, and saturation — Google's four golden signals, translated for a small team without dedicated SRE tooling.
Kubernetes Monitoring Without Prometheus and Grafana: What You Actually Need
Prometheus and Grafana are free to install and expensive to run. Here's what a small team actually needs from Kubernetes monitoring — and what you can skip.
Horizontal Pod Autoscaling: What Most Clusters Get Wrong
The majority of Kubernetes clusters run with no Horizontal Pod Autoscaler at all. Here's how HPA actually works and the mistakes that make teams turn it off.
Alert Fatigue Is Killing Your On-Call Rotation: A Practical Fix
Alert fatigue sets in within a single shift and burns out small teams fast. A concrete plan to cut noise and get back to a sustainable on-call rotation.
Auditing Kubernetes RBAC: Finding (and Fixing) Your ClusterRole Wildcards
A step-by-step walkthrough for auditing Kubernetes RBAC, finding overly broad ClusterRoleBindings, and replacing wildcards with least-privilege scopes.
The Real Cost of Kubernetes for a 5-Person Startup (And When It's Worth It)
Kubernetes isn't free even when the software is. An honest breakdown of engineering time, cloud spend, and hidden costs for a tiny team.
Namespace Strategy 101: Organizing a Cluster That Won't Turn Into a Mess
A practical Kubernetes namespace strategy for small teams — how to split by environment vs. team vs. service, and the naming conventions that scale.
Kubernetes Resource Requests and Limits: A Practical Right-Sizing Guide
How to set CPU and memory requests and limits that actually match real usage, instead of guessing and hoping — with a practical right-sizing workflow.
The Kubernetes Well-Architected Checklist: 12 Things to Check This Quarter
A practical, non-enterprise Kubernetes architecture checklist covering availability, security, cost, and operations — 12 checks to run this week.
SRE for Teams of One: Running On-Call Without a Dedicated Rotation
No dedicated SRE yet? Here's how to run a sane on-call rotation, set a sustainable alert budget, and avoid burning out your only backend engineer.