TL;DR: My main server file had grown to 2,426 lines. I used Claude Code to refactor it into clean modules, cutting 600 lines while keeping everything working in production. The Problem: When One...
TL;DR: The provisioning pipeline from Part 4 works great when everything goes right. This post is about the equally important code that runs when everything goes wrong. Last week was about the h...
TL;DR: No VM is created, resized, or modified until Stripe confirms payment. Everything else in the code exists to enforce that rule. Last week was about access boundaries — how users reach thei...
TL;DR: This platform gives users full SSH and console access to their lab VMs without ever exposing Proxmox, public IPs, or the underlying infrastructure. In Part 2, I laid out the architecture ...
In Part 1, I talked about why I built this platform. Not because I wanted to compete with hyperscalers — but because cloud KVM pricing simply doesn’t make sense when you want labs that stay up for...
Building a Budget-Friendly Lab VPS Platform Part 1 – The Idea ▶️ Watch the video If you’ve spent any amount of time labbing network automation, BGP, EVPN, or modern routing stacks, you’ve probably...
From Push to Proof: Drift validation and alerting with a Dispatcher-powered Interface Job Hook (Part 18) In the previous part, I used an interface-scoped Job Hook to enforce intent: render the int...
Nautobot Job Hooks - Reducing configuration drift between your live devices and your source of truth ▶️ Watch the video In this post, we’ll reduce configuration drift by wiring Nautobot J...
From Live Advisories to Persisted CVEs: Cisco PSIRT, Arista, and EoX This week we wire the pipeline into real vendor sources. You’ll see how the job authenticates to Cisco PSIRT, maps OS versions t...
Creating a Nautobot Job to manage vendor specific CVEs Building on last week’s foundation, this post dives deeper into the essential helpers for pulling platform data, NAPALM drivers, and IP addres...