FDA Clearance Is Not a Monitoring Plan
Clinical AI can fail without crashing. Physician-developers must build the monitoring, outcome linkage, and human checkpoints that keep cleared software safe after deployment.
· 18 min read
Surviving the AI-Native Transformation: A Physician-Developer's Guide
AI makes clinical software cheap to produce. It does not make it safe. Physician-developers must build systems in which speed remains subordinate to evidence, judgment, and the doctor-patient relationship.
· 6 min read
AI's Next Breakthrough Is Not a Bigger Brain. It Is a Memory You Can Trust.
The AI race has been about model size for years. The real bottleneck is memory: trustworthy, auditable knowledge that persists across years, not conversations. Medicine will feel this shift first.
· 10 min read
Patient Data Liberation Is Not Patient Data Protection
Wearables connected to medical records are not just a patient-access story. They are a boundary-design problem for doctors who code.
· 9 min read
Why Your Clinic Workstation Needs a GPU
A clinic workstation with local GPU inference changes the privacy, latency, and ownership posture of clinical AI workflows.
· 7 min read
Clinical AI Should Help Us Find the Growth Restriction Cases We Miss
Fetal growth restriction is not just an ultrasound diagnosis. It is a longitudinal data problem. Clinical AI will not replace MFM judgment, but it can help surface the pregnancies whose risk is already visible in the record.
· 9 min read
The Moment a Clinical Tool Becomes Infrastructure
A short introduction to a DoctorsWhoCode series on spreadsheets, tests, and retrieval as the basic discipline of physician-built clinical software.
· 7 min read
RAG Is the Bridge Between Medical Knowledge and Medical Practice
Clinical AI earns workflow trust only when its answers are grounded in current, local, auditable knowledge. Retrieval is not a feature. It is infrastructure.
· 9 min read
Dead Weight: What a 1970 Cardiology Textbook Taught Me About the Future of Clinical Evidence
A half-century-old textbook on congenital heart disease. A chapter on Ebstein's anomaly. A comparison that changed how I think about evidence, AI, and what we owe the next generation of physicians.
Fluent Answers Are Not Clinical Judgment
Language models can make uncertain medical information sound finished. The problem is not fluency. The problem is mistaking fluency for accountable clinical reasoning.
· 8 min read
Journals Are Becoming Infrastructure
Medical journals are no longer only read by clinicians. They are becoming upstream inputs to AI systems. That makes provenance, evidence hierarchy, and distribution part of clinical infrastructure.
· 8 min read
The Search Box Is Disappearing
Medicine is moving from retrieval to synthesis. That changes the physician's work from finding information to judging synthesized information under clinical pressure.
· 7 min read
AI Agents Do Not Replace Software Fundamentals. They Expose Whether You Have Any.
For physician-builders, agentic engineering is not prompt magic. It is bounded context, vertical slices, observability, and accountability.
· 8 min read
The Map Has Blind Spots
Move 37 was not a parlor trick. It was a warning. Physician-developers need to be ready for the moment AI starts finding medically important patterns our inherited maps never taught us to see.
· 8 min read
AI Just Outperformed Physicians at Clinical Writing. This Should Not Surprise You.
HealthBench Professional shows AI has already crossed the threshold in clinical writing and documentation. The real lesson is not replacement. It is that physician-developers need to build the harness.
· 8 min read
When AI Told a Dying Man What He Wanted to Hear
Joe Riley trusted AI over his oncologist and died of a treatable cancer. His tragedy wasn't naivety — it was earned distrust, amplified by a machine that had no way to know the difference.
The PDF Wall: Why Your Clinical RAG Keeps Hallucinating
Your clinical RAG system is not hallucinating because the model is bad. It is hallucinating because your document pipeline is broken. Here is what clinical PDF parsing actually requires, and why Docling is the fix.
· 11 min read
The Local Advantage: Why Physician-Developers Should Build on Local LLMs Instead of Consumer AI
Consumer AI tools like ChatGPT and Claude are useful. But a physician-developer who deploys a locally fine-tuned model on controlled infrastructure has something more powerful: a clinical tool that learns your practice, respects your data, and costs less at scale.
· 12 min read
Clinical Documentation Automation Should Remove Friction, Not Replace the Doctor
Documentation automation is not a typing solution. It is a workflow design problem. For physicians and physician-developers, the real goal is protecting clinical judgment from administrative drag.
· 8 min read
Doctors Who Code: Build Systems, Not Just Models
A TEDx pitch says physicians should build AI. I agree. But the work that matters is governance, validation, and delivery, not one-afternoon demos.
· 8 min read
The Referral System Is Broken for the Same Reason the Triage Line Is Broken
Nobody lost your fax. The system was designed to lose it. Referral failures are not clerical accidents. They are the predictable result of clinical infrastructure built for a different era.
· 12 min read
When Medical Algorithms Code Racism Into Patient Care
Race-based clinical algorithms in kidney care and obstetrics did not just reflect bias. They operationalized it. Physicians now have a responsibility to challenge the software, logic, and architecture that turn racial fiction into patient harm.
· 8 min read
Your Patients Are Already Using ChatGPT to Decide Whether to Call You
They are not asking it for fun. They are asking it because the triage line puts them on hold for 45 minutes. The threat is not the technology. The threat is the system that made the technology necessary.
· 10 min read
The EHR Vendor Wants You to Stay a Consultant
Physician passivity in health tech is not an accident. It is a business model. Understanding the structural incentives is the first step to building outside of them.
· 6 min read
The Limits of Viability: What Patients Find Before They Find You
When families face periviability, they search before they call. What they find shapes everything. Here is why physician-developers have a responsibility to build better.
· 7 min read
When the Algorithm Fails, Who Answers for It?
Every physician using an AI tool has heard the liability question. Most of us answer it wrong. The real answer is not about insurance. It is about who was in the room when the tool was designed.
· 6 min read
Augmented Intelligence Is a Physician Problem. That Makes It a Physician-Developer Opportunity.
The AMA opened the door. Physicians must decide what to do with it. The survey data is not a comfort. It's a challenge. Here's what physician-developers do next.
· 9 min read
Part 1: The $50 Billion Bait-and-Switch
Maternal-Fetal Medicine Specialist & Founder, CodeCraftMD
Part 2: What We Should Actually Build
Maternal-Fetal Medicine Specialist & Founder, CodeCraftMD
The Microphone Finally Got Smart: Dictation Before and After the LLM Era
Dragon NaturallySpeaking trained physicians to talk like machines. Ambient AI scribes are starting to listen like clinicians. Here is what changed after 2022 and why it matters for every doctor taking care of patients.
· 7 min read