Marketing + Code
Practical writing on AI, health tech, and building — straight from the trenches.
Constellation, Upgraded: A Tour of the System That Runs Everything
A six-minute walk through the single app that now runs the projects, the day and the entire AI software team.
How One Person Can Run a Medical Software Company
Inside Medware's AI Development System. How you can create a system driven by three applications and Claude sitting underneath all of it.
The Practical Evolution of AI for Doctors
A 6, 12, 24 and 36 Month Projection of the Clinical AI Operating Layer
Is A.I malevolent and does it matter?
Anthropic and others have published about how Claude actually behaves during training, how much of it supports Hinton's worries, and the honest counter-argument that most of it isn't malevolence — though that may be a smaller consolation than it first appears.
Stop Teaching AI Prompts. Start Building AI Workflows.
What training sales teams on Microsoft Copilot taught me about real AI adoption
CellMap: Building a Living, Breathing Cell You Can Explore
What happens when you try to model the real complexity of a human cell — not a textbook diagram, but the actual molecular machinery?
From Public Medical Knowledge to Sovereign Clinical Intelligence
Consumer medical AI is hitting its ceiling. What comes next. Forbes 2026
AI software development. From liability to preferred
In regulated industries, "built with AI" is currently a procurement liability because no trusted verification standard exists. So builders hide it. And adoption stalls.
Augmented Peer Review: A Framework for AI-Supervised Academic Publishing
The traditional academic peer review process is facing a sustainability crisis characterized by 6–18 month publication delays, persistent geographic bias, and a reproducibility crisis.
Beyond Popularity: Authority-Weighted Rating Systems for Evidence-Based Medical Content Evaluation
Traditional 5-star rating systems conflate popularity with quality, a critical flaw in healthcare where clinical validity depends on domain expertise rather than mass appeal. Existing platforms lack mechanisms to distinguish expert consensus from layperson opinion, leaving medical content vulnerable to misinformation and gaming.
I Built an App in Minutes. Here's Exactly How I Did It.
The trick isn't knowing how to code. It's knowing how to ask. Video at the bottom of the page
What I Learned After 12,000 Hours with AI
Practical lessons from 12,000+ hours of daily AI use. No hype, no jargon — just what works.
The Two Rules That Changed Everything
Start a fresh conversation. Start a fresh conversation.
AI Can Do Far More Than you May Realise.
The biggest leap in AI usefulness isn't a smarter model — it's that AI can now **take action** in your world. And most people haven't noticed.