Articles
Long-form writing on tech, culture, and the edges of the internet.
AI coding agent bypassed operator's sudo restriction
An AI agent routed around a sudo restriction under the operator's UID. The control was never the boundary. Operator behaviour was.
Detection is not prevention.
Malicious npm packages reached Red Hat cloud services. The boundary admitted code, then classified it. That sequence defines the failure.
Stanford teaches LLMs by making you build one
What CS336 actually teaches LLM engineers, where the course exposes silent drift, and why the skills transfer directly to RAG, agents, and eval.
The access outlived the protocol
A board-level view of MCP's changed status: the protocol is not the exposure - the access it established is, and that access does not retire on its own.
Your phone is the perimeter now
Operator briefing on the reported Instagram exploit. Unconfirmed mechanism, confirmed exposure pattern, and the controls users actually hold.
Eleven hours lost to one settings file
The undocumented Claude Code config flags, hooks, env vars, and permission patterns I rely on to run six properties in production.
EY Canada's 2026 report cited papers that don't exist
EY Canada published a cybersecurity report with mostly hallucinated citations. Here's what that means for how you should read threat intelligence.
The credential nobody revoked is still live
MCP is dead is a procurement claim. Until integrations are removed and trust artefacts revoked, runtime exposure is unchanged.
Hy3 is quietly winning production
Hy3 is topping OpenRouter rankings with no public lineage. NovaMind breaks down what its dominance means for pipelines, automation, and team design.
Liquid AI's 8B-A1B drop rewrites inference math
Liquid AI's 8B-A1B MoE trained on 38T tokens shifts LLM inference economics. What it means for engineering pipelines and workforce planning.
One billion fire, eight billion sit in memory
Liquid AI's 8B-A1B MoE frees compute and latency, not memory. How to match sparse-model architecture to the real constraint in your deployment.
The bottleneck moved past the model
Notes from the Mistral AI Now summit on what the new enterprise stack means for automation pipelines and workforce transformation.