Writing Hub
AI governance essays, reasoning systems notes, experiment logs, and technical writing across BioAI and engineering practice.
Project Topics
We Built AI Verification Infrastructure. Then It Found Our Blind Spots.
A technical account of the Flamehaven Verification Ledger — what it found, where it failed, and what we need the field to tell us
Stanford. Princeton. A bioRxiv Paper. So Why Did Nobody Ask Where the Data Goes?
BioClaw processes EHR data. Its primary showcase channel is WhatsApp. We audited the repository: 60/100, Tier 2 Caution. Here is what the bioRxiv paper says that the README does not.
From Repo Scanner to Audit Architecture: What Changed in STEM BIO-AI Through v1.7.8
A technical look at how STEM BIO-AI v1.7.8 became less Python-shaped, more semantically stable, and more inspectable across real audit output surfaces.
How Do You Trust the AI Auditor? STEM-AI v1.1.2 and Memory-Contracted Bio-AI Audits
STEM-AI v1.1.2 binds a bio/medical AI repository audit to a machine-checkable memory contract, then demonstrates it on a real open-source bioinformatics repository.
The $100 Million Blind Spot: What No-Code Healthcare Builders Still Don't See
An analysis of how no-code and AI-generated healthcare apps create regulatory liability when patient data flows are deployed without prior mapping, auditability, or compliance architecture.
Bio-AI Repository Audit 2026: A Technical Report on 10 Open-Source Systems
We audited 10 prominent open-source Bio-AI repositories using code inspection and STEM-AI trust scoring. 8 of 10 scored T0: trust not established. Here is what the code actually shows.
How do you know when your entire AI pipeline is wrong — not just one model? (EXP-033)
EXP-033 shows how to validate an entire AI pipeline, not just one model, using five-gate checkpoints, reproducible PASS/BLOCK parity, AlphaGenome on/off testing, and fully traceable governance decisions.
What AI Changed About Research Code — and What It Didn’t
The old bottleneck was writing the code. The new bottleneck is proving that the code still means what the theory meant.
What an AI Reasoning Engine Built for Alzheimer's Metabolic Research: A Code Walkthrough
A code walkthrough of an AI reasoning engine for Alzheimer’s metabolic research, showing how literature ingestion, causal inference, and executable biomarker scaffolds generate falsifiable pre-validation hypotheses.
Chaos Engineering for AI: Validating a Fail-Closed Pipeline with Fake Data and Math
A case study in AI governance showing how synthetic invalid inputs, structural disagreement, SIDRCE ethics checks, and end-to-end reliability scoring triggered a safe BLOCK verdict in a biomedical pipeline.
When AI Models Fight, Truth Wins: The “Eureka” Moment for Tired Researchers
To the grad student staring at a pLDDT of 90 and wondering why the ligand won’t bind.
From 97% Model Accuracy to 74% Clinical Reliability: Building RSN-NNSL-GATE-001
Learn how RSN-NNSL-GATE-001 turns high model accuracy into system-level clinical reliability by blocking unsafe AI pipeline decisions, measuring end-to-end risk, and enforcing fail-closed governance.
Showing page 1 of 2 · 16 matching posts