Writing Hub
AI governance essays, reasoning systems notes, experiment logs, and technical writing across BioAI and engineering practice.
Project Topics
AI Agents Are Poisoning Your Codebase From the Inside
Explore how AI-generated code can silently degrade software quality through weakened tests, rising code churn, and duplication—and how teams can prevent it with better governance.
Implementing "Refusal-First" RAG: Why We Architected Our AI to Say 'I Don't Know'
Implementing refusal-first RAG means teaching AI to say “I don’t know.” This article explains evidence atomization, Slop Gates, and grounding checks that favor verifiable answers over plausible hallucinations.
LOGOS LawBinder: From Governed Reasoning to Audit-Grade Execution
This article explains how LOGOS v1.4.1 improves production AI reasoning with multi-engine orchestration, complexity-aware governance, and audit-friendly failure tracing.
LOGOS v1.4.1: Building Multi-Engine AI Reasoning You Can Actually Trust
LOGOS v1.4.1 is a multi-engine AI reasoning orchestrator that enforces consensus, traces failures, and applies governance profiles to reduce drift and make production reasoning more trustworthy.
LawBinder v1.3.0: Governance as a Kernel (Not a Guardrail)
LawBinder v1.3.0 shows how AI governance can run like a kernel, using deterministic Rust-based enforcement, replayable audit signatures, and bounded-latency policy checks in the critical path.
Why I Stopped Treating Complexity as a Bug
On intent, governance, and why “clean code” heuristics fail in AI-generated systems
The Real Risk in the Age of AI Coding Isn’t Bugs
Is your AI code production-ready or just 'AI Slop'? Learn how to detect convincingly empty code, measure Logic Density (LDR), and stop 'Vibe Coding' from becoming hidden technical debt.
Turning a Research Paper into a Runnable System
Turn a research paper into a runnable system. This article shows how HRPO’s core equations were implemented with bounded policy lag, KL rejection, and execution checks to test real-world fidelity.
Why AI Dismisses Your Best Work in One Second
Why do AI models dismiss original work in seconds? This essay explores the hidden mechanics of AI skimming—shortcut learning, probabilistic safety, fast-thinking defaults, and why depth requires time.
My Code Fixed Itself at 11PM
A “Quantum Engine” is a dramatic name. Here’s the un-dramatic story.
⌨️She Said I Broke the Speed of Light. So I Turned It Into Math.
"You broke the speed of light." Instead of fighting the cosmic war, I built the Composite Reliability Index (CRI). It's the engineer's math model for filtering online noise, saving your sanity, and reclaiming your afternoons.
Running the “Anti-AI” Playbook Through the Debugger
Critics say AI is broken — hallucinations, hype, and no ROI. But what if those bugs aren’t failures, but blueprints? This article runs the 10 most common anti-AI arguments through the debugger to reveal what’s really coming in Gen-2 AI.
Showing page 3 of 4 · 45 matching posts