It started with the humble intention of making my company issued Windows laptop just a bit more secure and private. But my unassuming, minor configuration tweaks and small scripts needed constant improvements, and I soon found myself spending increasingly less time playing with the dog and more time at my desk, perfecting responsive tasks, super granular-permissioned processes, ephemeral VMs and sandboxes, honeypots, and making countless changes to user accounts and monitoring jobs. Gradually, my PC had evolved into a Rube Goldberg’s machine of defenses. Everything worked, but felt overly exaggerated and complicated. Tired of playing catch-up with Microsoft and with new attack tactics, I eventually switched to Qubes OS. This is my attempt at simple desktop security using basic tools and scripts, hoping to create defenses that would be straight-forward to understand and could be easily audited and replicated. I’ll explain what went well and what didn’t, discuss tricks and “gotcha” moments and things that I’d avoid if I was to repeat the effort. The second part of the talk will cover Qubes OS, "a reasonably secure operating system", which achieves isolation through multiple Xen-powered environments. I’ll go over my experience switching to Qubes OS, what I liked and what could be improved, and whether or not a system like Qubes can be used organizationally at scale.