Engineering Devlog
Infrastructure, GitOps & Autonomous Robotics
Hands-on technical logs chronicling the construction of a fully reproducible bare-metal Kubernetes platform on Proxmox, scaling to an Argo Workflows + Gazebo headless simulation farm.
π₯οΈ Active Cluster
3x Proxmox VE 9 Nodes
40 CPU Cores • 157 GB RAM
RTX 3090 GPU Passthrough
40 CPU Cores • 157 GB RAM
RTX 3090 GPU Passthrough
π Recent Engineering Logs
View Full Roadmap →Phase 5
October 28, 2026
15 min readPhase 5, Part 1: Prometheus, PromQL and My First SLO Alert
Installing kube-prometheus-stack, learning PromQL from the `up` metric through histograms, then probing Grafana with blackbox exporter and wiring a multi-window burn-rate SLO alert from the Google SRE Workbook.
Phase 2
September 29, 2026
6 min readPhase 2, Part 1: cert-manager, Rate Limits, and the Restore Race
Handling cert-manager Helm installation, Let's Encrypt DNS-01 wildcard certificates, rate limiting, and bootstrap secret timing races during full teardowns.
Phase 0
September 22, 2026
7 min readPhase 0, Part 1: Turning Copy-Pasted Terraform Into a Reusable Proxmox Module
Refactoring duplicated Proxmox VM and LXC declarations into a shared module, centralizing network outputs, and using git subtree for offline disaster recovery.