RC RANDOM CHAOS

Articles

Long-form writing on tech, culture, and the edges of the internet.

Stanford teaches LLMs by making you build one
LLM engineeringCS336

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.

9 min read
The access outlived the protocol
MCPboard governance

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.

7 min read
Your phone is the perimeter now
instagram securityaccount compromise

Your phone is the perimeter now

Operator briefing on the reported Instagram exploit. Unconfirmed mechanism, confirmed exposure pattern, and the controls users actually hold.

7 min read
Eleven hours lost to one settings file
claude-codeanthropic

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.

5 min read
EY Canada's 2026 report cited papers that don't exist
threat intelligenceAI safety

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.

7 min read
The credential nobody revoked is still live
MCPprotocol deprecation

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.

8 min read
Hy3 is quietly winning production
Hy3 LLMOpenRouter

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.

8 min read
Liquid AI's 8B-A1B drop rewrites inference math
mixture of expertsLLM engineering

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.

9 min read
One billion fire, eight billion sit in memory
MoE architectureLLM engineering

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.

9 min read
The bottleneck moved past the model
Mistral AIworkforce transformation

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.

9 min read
800 servers gone, the scans kept coming
AS209847FIOD seizure

800 servers gone, the scans kept coming

Dutch FIOD seized 800 servers from AS209847. One week later the scan rate is unchanged. What that signal actually means.

5 min read
One PIN unlocks the vault
vaultjackinggoogle password manager

One PIN unlocks the vault

Vaultjacking turns one captured PIN into full retrieval of a Google Password Manager vault. Operator breakdown of the control collapse.

7 min read