Identity · Security · AI infrastructure
I write down the things I learn, usually after I've made the mistake first. Long-form posts on what I'm building, breaking, and defending.
Running Wan 2.2 image-to-video on a DGX Spark to make a hidden bedroom scene breathe. The Blackwell setup traps nobody documented yet, the model that kept summoning a person I never asked for, and the trick that puts live text on a monitor inside the video.
Claude Fable 5 is genuinely the most capable model Anthropic has shipped. It also blocks legit security work, gates the uncapped version behind an invite list, and quietly drops the zero-retention floor a lot of business customers chose Claude for.
A pocket travel router with an IPsec tunnel back to a network I actually control, broadcasting a dedicated SSID that follows me into every hotel, airport, and coffee shop. Untrusted wifi stops being a thing I have to think about. And by total coincidence, the Chromecast at home thinks it lives somewhere it can watch what I pay for.
Claude Opus 4.8 is a sharpening, not a leap. Clear gains, more honesty, hundreds of parallel subagents, and what it all means for security teams.
A hands-on first-hours review of Claude Opus 4.7. What moved, what broke, the token-cost gotchas, and why Anthropic is still holding Mythos back.
A sharp take on why today’s AI agents feel wrong, and what a quieter, more persistent AI-native future could actually look like.
Troubleshoot Nvidia DGX Spark setup issues and improve your AI Workbench custom container with these step-by-step solutions for a usable environment
