Kubernetes, Cloud Native & AI Research, Consulting, and Training
CNCF Ambassador / Kubestronaut / AIOps Researcher
Port an MCP server to the stateless spec on Kubernetes: a step-by-step migration
Porting the transport to the v2 SDK is mostly a version bump. The real migration is deciding where handle state lives, and leaving it in pod memory brings the old failure back behind an HTTP 200 that transport retries never see.
How much CPU and memory does a CNI actually use?
A CNI is picked once and rarely revisited, and there is no organized data on what it consumes day to day. Measuring 5 CNIs in 14 configurations under identical conditions showed an 8x memory spread, an eBPF map share that kubectl top never sees, and a 70% kube-proxy memory saving from the nftables mode that had not been published as a number.
Does switching CLAUDE.md to AGENTS.md slow Claude Code down?
The two AGENTS.md migration workarounds (import, symlink) were measured on Kubernetes incident-response tasks: no penalty in speed or context-load tokens; deltas moved with the symlink control, marking them as run variance.
Passing Conformance Isn't Enough: Re-scoring 7 Gateway API Implementations from Scratch
We rebuilt last year’s Gateway API PoC from the scoring model up and re-measured 7 implementations. All 7 pass official conformance, but the features you can actually use range from 6 to 13.
How well do Claude / Gemini / Codex CLI agents handle Kubernetes incidents? (AIOps Agent Benchmark)
Putting nine agents through ten Kubernetes incident scenarios under identical conditions surfaced two things: the Efficient tier beat Flagship outright, and each brand has a clearly different operational personality.
cc-deck: Claude Code Session Hub
A cross-platform TUI tool that makes Claude Code sessions searchable and manageable: fuzzy search, TODO auto-pinning, manual bookmarks, quick queries, size management for oversized sessions, and smart resume.
49% Cost Reduction on a Large EKS Cluster: Clearing Infrastructure Debt by Decommissioning V1 and Rebuilding V2
~$3,240/month, ~$38,880/year saved (~49%). Consolidated EKS 1.32/1.33 to 1.35, ArgoCD v1 to v3, and Kafka Zookeeper to KRaft in one shot while reducing nodes from 25 to 15. Achieved fully zero-downtime cutover with Route53 Weighted Routing.
Cutting a Small Kubernetes Cluster Bill by 48%: A Dual-Cloud EKS + AKS Optimization Story
~$414/mo, ~$5,000/yr saved combined (~48%): 39% on production, 61% on dev. Six waste patterns and a zero-downtime execution playbook for small K8s clusters.
Comparing 7 Kubernetes Gateway API Implementations: Which One Should You Choose?
100-round PoC results and selection guide for 7 Kubernetes Gateway API implementations across 17 test scenarios