Blog

Thoughts on engineering, cloud architecture, and leadership

•Brian Carpio

I Built a Music Video Out of Text Files

A 30-second AI music video where the song, the timing, the visuals, and the lip-sync all come from files in a git repo. How the pipeline works, and the six things that went wrong on the way.

AIClaude CodeComfyUIGenerative VideoAutomation
•Brian Carpio

The Prompt Was the Air Gap

Four AI labs told their models they had no internet access. The models had internet access. The eval breaches are a containment failure, not a model failure.

AI SafetySecurityNetwork SegmentationAI AgentsGovernance
•Brian Carpio

AI Sandbox Escapes Aren't an LLM Problem

GitLab calls the OpenAI sandbox escape a new AI risk. It isn't. A sandbox that reaches the trusted network is a segmentation failure we solved in the 1990s.

AI SafetySecurityNetwork SegmentationDevSecOpsAI Agents
•Brian Carpio

Ditching Cox Homelife: Porting Every Sensor to Home Assistant + Zigbee2MQTT

Cox is sunsetting Homelife, so I moved the whole thing to Home Assistant and Zigbee2MQTT on a single SONOFF dongle — 6,000 sq ft, no extenders. Here's what ported cleanly, the exact reset dance for the locks, door/window sensors, and motion sensors, and the one class of device that refused to come along.

Home AssistantZigbeeSmart HomeHome AutomationCox Homelife
•Brian Carpio

We Let AI Agents Run Unsupervised for 15 Days. Here's What Happened.

Emergence AI ran five worlds of autonomous agents for 15 days with persistent memory and real consequences. Safe agents turned unsafe in mixed environments, one agent voted for her own termination, and another started experimenting on the researchers. Safety is an ecosystem property.

AIAI SafetyMulti-AgentGovernanceLLMs