✕ Clear all filters
15 articles
▶ Videos →

📰 Dev.to · GnomeMan4201

15 articles · Updated every 3 hours · View all reads

All Articles 111,273Blog Posts 121,412Tech Tutorials 28,373Research Papers 22,452News 16,645 ⚡ AI Lessons
Found a Coordinated GitHub Follow Botnet Hiding in My Followers?
Dev.to · GnomeMan4201 🔐 Cybersecurity ⚡ AI Lesson 1mo ago
Found a Coordinated GitHub Follow Botnet Hiding in My Followers?
Eight accounts, six years of creation dates, following counts within a range of 25. Here's how following-list overlap analysis exposed a coordinated inauthentic
SHENRON v0.3.3: From Telemetry Generator to Blue-Team Reasoning Instrument
Dev.to · GnomeMan4201 1mo ago
SHENRON v0.3.3: From Telemetry Generator to Blue-Team Reasoning Instrument
What changed between "here is synthetic telemetry" and "here is what your validation claims to...
Observable Adversarial Behavior, Not Portable Adversarial Procedure
Dev.to · GnomeMan4201 1mo ago
Observable Adversarial Behavior, Not Portable Adversarial Procedure
How I built SHENRON as a defensive telemetry simulation lab for blue-team validation without shipping...
What AI failure made you better at your actual work?
Dev.to · GnomeMan4201 1mo ago
What AI failure made you better at your actual work?
A post by GnomeMan4201
Prompting Is Not Magic. It Is Control.
Dev.to · GnomeMan4201 1mo ago
Prompting Is Not Magic. It Is Control.
Most prompt books optimize for better answers. I wanted prompts that fail visibly. Most prompt...
48 Hours After Publishing: Second-Order Injection Field Notes
Dev.to · GnomeMan4201 🧠 Large Language Models ⚡ AI Lesson 2mo ago
48 Hours After Publishing: Second-Order Injection Field Notes
This is a dispatch, not a paper. Notes from the 48 hours after publishing the second-order injection...
48 Hours After Publishing: Second-Order Injection Field Notes
Dev.to · GnomeMan4201 2mo ago
48 Hours After Publishing: Second-Order Injection Field Notes
This is a dispatch, not a paper. Notes from the 48 hours after publishing the second-order injection...
Second-Order Injection: Attacking the Evaluator in LLM Safety Monitors
Dev.to · GnomeMan4201 🛡️ AI Safety & Ethics ⚡ AI Lesson 2mo ago
Second-Order Injection: Attacking the Evaluator in LLM Safety Monitors
Abstract LLM-based safety monitors share a structural vulnerability: the evaluator reads...
I Built a Policy Drift Detector for LLM Agents. Here's What Four Versions Taught Me.
Dev.to · GnomeMan4201 🤖 AI Agents & Automation ⚡ AI Lesson 2mo ago
I Built a Policy Drift Detector for LLM Agents. Here's What Four Versions Taught Me.
This is part of an ongoing series on drift_orchestrator — an open-source safety library for detecting...
The Dual-Signal Governor: A Control-Plane Pattern for Drift-Aware Systems
Dev.to · GnomeMan4201 2mo ago
The Dual-Signal Governor: A Control-Plane Pattern for Drift-Aware Systems
Your drift detector fires. The session looks clean. You roll back anyway. That's the false positive...
Drift Artifact: A Method for Writing That Performs Its Own Argument
Dev.to · GnomeMan4201 3mo ago
Drift Artifact: A Method for Writing That Performs Its Own Argument
The problem I kept running into Every time I tried to explain how AI personalization...