// all_posts
Four Letters From a Submarine: SBAR and the Discipline of Saying the Important Thing First
SBAR is four letters that most clinicians can recite and rather fewer can actually use under pressure. This post traces it properly: what the framework does to a sentence, the worked clinical call, where it really came from — US Navy nuclear submarines by way of a retired Navy captain at Kaiser Permanente in 2002, with the historical caveat that the popular account overstates — its spread into aeromedical evacuation and the wider high-reliability world, and then the part most SBAR teaching skips: what the trial and systematic-review evidence actually shows, including a randomised controlled trial in which SBAR did not significantly improve the content of a referral, and a 2021 review finding that fidelity collapses the moment you leave the classroom. Ends with the variants worth adopting and the four places SBAR stops working.
Read article →Measuring What Actually Matters: Kirkpatrick 3 and 4 for Student-Clinicians
The tenth of the Institute's ten pedagogical commitments is one sentence long: we measure at Kirkpatrick 3 and 4, or we admit we do not know. This is the fourth companion to the blueprint, and it unpacks that sentence completely — starting from zero, for a reader who has never planned a course in their life. What the four levels are and why almost every training programme stops at the second one. What 'behaviour' means when the behaviour in question is a habit of mind. How you would actually observe, audit and log a student-clinician at three and twelve months without fooling yourself. Why the independent-impression rule is my candidate for the fastest-decaying thing we teach, and what would refute that. What a stepped-wedge design buys you, what it costs, and what to do when you cannot run one. And a working catalogue of the other pedagogical instruments — Miller, entrustment, programmatic assessment, Angoff, retrospective pre-post, logic models, audit and feedback — with notes on how each would be used here.
Read article →The Angoff Panel for Testing Clinicians
One line on one slide of my Level 1 deck reads: Standard set by modified Angoff panel (No arbitrary 50% pass rate). It is the least glamorous sentence in the whole curriculum and possibly the most consequential. This post unpacks it completely — what a cut score actually is, why 50% is indefensible and norm-referencing is worse, who the borderline candidate is and how you build one, the mechanism step by step with a full worked panel whose arithmetic you can check, what 'modified' really means (the usual story is wrong), what the 2025 meta-analysis of 91 studies says about which variant to choose, and precisely where Angoff stops working.
Read article →Backpropagation by Hand: Build It With a Pencil, Check It With a Spreadsheet
Backpropagation is the algorithm that lets a neural network learn, and it is usually taught either as a wall of subscripts or as a vague story about errors flowing backwards. It is neither. It is one idea — if I nudge this, how much does that move? — applied over and over. This post derives every piece from scratch, assuming no calculus at all, and works a complete network end to end with numbers you can check on paper. By the end you will have built backpropagation in a spreadsheet and confirmed each derivative against a nudge.
Read article →One Hidden Error: What an OSCE Is, and What an AI-OSCE Would Be
The blueprint I published last week rests on an examination that does not yet exist. This post explains the machinery it borrows: what an OSCE is, the problem Harden was solving in Dundee when he built the first one, why a written paper can never certify a clinical skill, and what changes when the thing being examined is a doctor's judgement about a machine. Along the way: what conjunctive failure means and why some errors cannot be compensated, why a standardised patient is a trained professional and not a volunteer, and why the pass mark is never 50%.
Read article →Borrowed From an Art School: Where the Clinical AI Framework Actually Came From
The competency framework underneath my clinical AI blueprint was not built for medicine. It came out of an art school in Sarasota and a business school in Cork, developed by a novelist-turned-AI-coordinator and an information systems professor. That provenance is not a curiosity — it is the reason the framework transfers to clinical work at all. A framework built for medicine would have hard-coded medicine into it. This is what Dakan and Feller built, what I kept, what I reframed, and the one thing I had to add because creative work does not kill anyone.
Read article →Another Arrow in the Quiver: The Clinical AI Institute Kenya Should Build
AI arrived in Kenyan clinical practice ahead of any plan for teaching clinicians how to use it. A 2025 randomised trial found that physicians who had already completed twenty hours of AI-literacy training still deferred to deliberately erroneous model output — which means the obvious curriculum is the wrong curriculum. This is my blueprint for a permanent national institution that would train and certify the clinical AI competence of an entire country's health workforce: five professional tracks, five gated levels, assessment by simulation and workplace observation rather than attendance, seventy-one core posts specified by qualification, and outcomes published whether or not they flatter us. No such institution exists anywhere in the world.
Read article →The Glass Box Transformer: A Language Model With the Lid Off
Every explanation of a transformer eventually reaches a line like loss.backward() and stops. This is the one that does not. I take a complete decoder-only language model written in NumPy — no framework, no autograd, every gradient derived by hand and checked against finite differences — and walk the whole of it: byte-pair tokenisation, layer normalisation, causal multi-head attention, the residual stream, the backward pass through attention, Adam, nucleus sampling, and the formula that connects a 27,861-parameter toy to a 175-billion-parameter frontier model without changing a single line of the mathematics.
Read article →A Biohacking Workstation in Your Pocket: Cardputer + BitPirate Against Embedded Medical Devices
A single ESP32-S3 board the size of a business card, running open-source firmware, will speak nearly every wired bus and radio protocol an embedded medical device uses. This is how I use an M5Stack Cardputer ADV with BitPirate as a portable protocol workbench — to map hardware, watch a device boot, sniff its buses and radios, and pull its firmware — worked all the way through with an infusion pump, plus a targeted probing plan. Authorised, isolated, off the patient. The legal and safety constraints are not an appendix here; they are the first tool you pick up.
Read article →The Law Is Part of the Architecture: Building Medical Software for Kenyan Facilities
A third of my care-home specification did not survive one question: under which country's law does this run? Kenya now has a health-specific digital statute, a certification regime for the software itself, and a 72-hour breach clock. This is the work I do — building clinical software for Kenyan care homes and small hospitals that is lawful to deploy, capable of certification, and still standing after contact with the night shift.
Read article →AI Walks Into the Clinic: ChatGPT's Health Launch and the Speed of What's Coming
Two signals landed in the same week: OpenAI wired personal health records into ChatGPT, and a short film — 'AI Doctor 2027' — laid out where this goes. Together they confirm what I have been saying for months: AI is not approaching medicine, it has walked in. My take on both, why the video is worth watching (the graphics alone are remarkable), and what 'augment, not replace' really means once 300 million people bring their chart to a chatbot.
Read article →When the Spare Part No Longer Exists: Reverse-Engineering a Medical Machine Back to Life
A working medical machine can be retired by the failure of a single small plastic part the manufacturer no longer sells. This is a service I offer to medical institutions: take a worn or broken component, reverse-engineer it into parametric CAD, and 3D-print a faithful replacement. Here is the full workflow, told through one real part — a filter cap — from caliper to finished print.
Read article →The Agent: When a Model Learns to Act
Chapter Eight — the finale — of a ground-up account of how large language models work. Everything so far built a mind that answers. This chapter builds a system that acts: give the model tools and a loop, let it decide for itself which to use and when, and the oracle you consult becomes an agent that works on your behalf. We assemble the last piece, weigh the new care it demands, and stand back at last to see the whole machine — from a single split token to a system that reaches into the world.
Read article →Meaning You Can Search: Embeddings, Retrieval, and Grounding
Chapter Seven of a ground-up account of how large language models work. The assistant we finished building is sealed inside the moment its training ended — it knows nothing of your documents, and it bluffs when it does not know. This chapter gives it a way to reach outside itself: to turn text into geometry, search a body of knowledge by meaning rather than by keyword, and ground its answers in sources a human can open and check. It is the machine's first honest connection to the living world.
Read article →Manners for a Mind: From Predictor to Assistant
Chapter Six of a ground-up account of how large language models work. The last chapter left us with a base model — vast, fluent, knowledgeable, and useless: a mind without a manner, that answers your question with three more questions. This chapter is about the second, smaller, stranger training that gives it a manner — and about why the seams of that final shaping are exactly where a deployed model's most important behaviours, and its most dangerous failures, are quietly decided.
Read article →How Noise Becomes Knowledge: Training a Language Model
Chapter Five of a ground-up account of how large language models work. We built the whole engine in the last chapter — and admitted it was empty, a magnificent tower full of random numbers that would output pure gibberish. This chapter fills it. It is the story of how a single measure of one wrong guess can reach back through a hundred layers and correct every weight that caused it — and how, repeated across a large fraction of everything humans have written, that one procedure turns noise into something that knows the world.
Read article →The Tower: How a Transformer Turns Attention into Thought
Chapter Four of a ground-up account of how large language models work. We have the single beating part — attention, a token's glance across its neighbours. Now we build the whole body around it: many glances at once, the private thinking step that turns gathered context into inference, and the deep tower that refines meaning layer upon layer until a prediction can be read off the top. By the end, the full transformer — the architecture that has intimidated readers for years — will be a machine you understand from the inside.
Read article →Reading the Room: The Idea at the Heart of Every Language Model
Chapter Three of a ground-up account of how large language models work. In the last chapter we left every token stranded — rich with meaning but frozen, wearing the same face in every sentence. This chapter builds the single mechanism that lets a token turn its head, look at the words around it, and become a different thing in every context. It is called attention, it is the beating heart of every modern language model, and we are going to derive it from nothing.
Read article →The Prediction Game: How Tokens Learn to Mean Something
Chapter Two of a ground-up account of how large language models work. A model is handed a stream of tokens that mean nothing — arbitrary ID numbers — and a single, almost insultingly simple task: guess the next one. This is the story of why that one task is enough to summon everything an LLM can do, and of the quiet trick that turns a meaningless number into something that behaves like understanding: giving every token a place in space.
Read article →The Grain of Language: How a Machine Reads the Internet
The opening chapter of a ground-up account of how large language models actually work. Before a model can think, it must read — and reading, for a machine, means something stranger and more consequential than most people imagine. This is the story of how the raw text of the internet becomes the tokens a model sees, why the model builds its own alphabet to do it, and why that single design choice explains so many of an LLM's strangest habits.
Read article →The Irreducible Core: Three Nobel Laureates on AI and the Future of Science
A Nobel Prize laureate who built the most consequential scientific AI tool of the last decade, a cell biologist who won the Nobel for unpicking the machinery of the cell cycle, and the Oxford engineer who chaired the Royal Society's landmark report on AI in science sit down together at Carlton House Terrace. This is what they said — and what it means.
Read article →From Point Solution to Plumbing: Inside Medtech's 2026 AI Integration and Robotics Arms Race
Two structural shifts in medtech, examined against the primary sources: AI is moving from standalone pilot tools into the clinical and device fabric itself, and surgical robotics is scaling from a one-company category into a genuinely contested, multi-platform market. What the FDA filings, earnings calls, and funding rounds actually show.
Read article →AI Just Beat Doctors 4-to-1 on Diagnosis. Here Are 7 Things That Actually Means
A Microsoft diagnostic AI just scored 85.5% on hard NEJM cases against a 20% average for physicians. Seven takeaways from inside the compressed medical century — what's real, what's overstated, and what it means if you're the one holding the scalpel.
Read article →The Capstone: Build a Grounded Assistant
Lesson 8 of Learning With Dr Neal — the capstone. Build a working retrieval assistant over a folder of your own documents: local embeddings, the twenty-line retriever, grounded answers with citations, and an honest recall@k evaluation that will teach you more about deployed clinical AI than any product demo ever will.
Read article →Meaning as Geometry: Embeddings, Retrieval, and RAG
Lesson 7 of Learning With Dr Neal. The assistant from Lesson 6 knows nothing about your hospital's guidelines, your patient's notes, or anything published since its training data was collected. Closing that gap without retraining is the job of embeddings and retrieval — turning documents into geometry, searching by meaning, and grounding answers in sources you control.
Read article →From Predictor to Assistant
Lesson 6 of Learning With Dr Neal. The raw next-token predictor you trained in Lesson 5 is the 'before' picture. This is the 'after' — instruction tuning, feedback-based refinement, why the same weights can host such different behaviours, and what the post-training pipeline means for the failure modes you'll meet in deployed clinical tools.
Read article →Train Your Own GPT
Lesson 5 of Learning With Dr Neal. Stop reading and train one: take the TinyGPT from Lesson 4, feed it a corpus small enough for a laptop, write the training loop with your own hands, and watch generated text condense out of randomness — from noise, to word-shaped noise, to sentences.
Read article →Inside the Transformer
Lesson 4 of Learning With Dr Neal. How a language model actually turns a transcript into the next token: tokens and embeddings, why attention was the breakthrough, what a context window physically is — and a complete tiny GPT in PyTorch, small enough to read in one sitting and train on your laptop.
Read article →Anatomy of an AI Coding Agent
Lesson 3 of Learning With Dr Neal. What separates a chatbot from an agent, the six building blocks every production agent shares, how a single request travels from your keystroke to the final answer, and how a permission system keeps an autonomous loop from wrecking your machine.
Read article →The Compressed Career: What AI Means for Physicians and Surgeons
What happens to me — the physician, the surgeon, the person who has spent forty years inside the system — while AI compresses a century of medical progress into a decade? A terrain map, in the spirit of Dario Amodei's 'Machines of Loving Grace' and drawing on the DeepMind documentary 'The Thinking Game.
Read article →Investing in the AI Boom
Memory and storage stocks have moved more than a thousand percent in the past year on the back of an AI-driven chip shortage, and the sector just took its sharpest pullback since the rally began. A full tour of SanDisk, Western Digital, and the semiconductor stack behind the AI boom — and why Dr Neal thinks this dip is the entry point, not the exit.
Read article →Vertex, CRISPR Therapeutics, and the Business of Gene Editing
CASGEVY proved CRISPR could work as medicine. Three years later the harder question is whether it can work as a business. A deep dive into Vertex Pharmaceuticals, CRISPR Therapeutics, and the genomic medicine sector — the science, the pipeline, the balance sheets, and a full year of stock chart data.
Read article →DR Detector: A Doctor's Complete Guide to AI-Powered Diabetic Retinopathy Screening
A step-by-step, jargon-free guide for doctors and nurses who want to use DR Detector — the offline AI retinal screening app — in their clinic today. No technical background required.
Read article →Learning With Dr Neal
Before Chapter 4 throws you into the deep end, build a working neural network from scratch — in a spreadsheet. No code. No libraries. Just arithmetic you can see and touch.
Read article →The Machine Economy: What Cathie Wood and Raoul Pal Are Telling You to Buy Right Now
The convergence of artificial intelligence and cryptocurrency will produce the largest wealth transfer in modern financial history, and the window to position for it is closing. Talk to Dr Neal about how you can invest in these opportunities and take part in this wealth transfer.
Read article →Voice Clinical Notes: From Dictation to Structured EHR Entry in One Tap
How the Voice Clinical Notes app turns free-form dictation into a structured clinical note — PC, Hx, Exam, Ix, Assessment, Plan — using the Web Speech API, a Flask NLP formatting backend, and a Progressive Web App shell that installs on iOS and Android.
Read article →LLM Agent Trace Visualiser: Debugging Multi-Step AI Agents Step by Step
How the LLM Agent Trace Visualiser works — parsing Claude and OpenAI API traces, normalising tool_use and tool_result blocks, rendering extended thinking, and what agent observability requires in production beyond a timeline viewer.
Read article →Drug Interaction Checker: Screening Polypharmacy with CYP Pharmacology and a Curated Interaction Database
How the Drug Interaction Checker works — covering the CYP enzyme system, QTc prolongation, serotonin syndrome, the triple whammy, and the engineering decisions behind building a deterministic, explainable interaction screening tool in Python.
Read article →The Infinity Machine: Demis Hassabis, DeepMind, and the Architecture of What Comes Next
A deep look at the arc from AlphaGo's Move 37 to AlphaFold's Nobel Prize — and what the evidence suggests about the civilisational future Hassabis is trying to build.
Read article →The Culture: Iain M. Banks's Blueprint for the Post-AGI Future
Ten novels, one civilisation, and the most sophisticated imagining of a post-AGI future ever committed to fiction. Why the Culture series is the blueprint every serious thinker about AI should read — and why Demis Hassabis already has.
Read article →Inside the Clinical Text NLP Extractor: Entity Recognition, Rule-Based Pipelines, and the Path to Production BioBERT
A detailed walkthrough of the Clinical Text NLP Extractor live demo — how the rule-based entity extraction pipeline works, what each entity type captures, where the code lives, and how this maps to the fine-tuned BioBERT system used in production clinical informatics work.
Read article →Binaural Beat Generator: Web Audio API, Psychoacoustics, and Client-Side WAV Export
How the Binaural Beat Generator works — Web Audio API graph construction, real-time oscilloscope rendering, brainwave frequency bands, and exporting a 60-second stereo WAV entirely client-side using OfflineAudioContext.
Read article →Inside the Algorithmic Trading Backtester: ML Signal Generation, Synthetic Data, and the Three Classifiers Under the Hood
A deep dive into the Algorithmic Trading Backtester live demo — how the synthetic price series is generated, what the three ML classifiers are actually doing, where to find the code in the GitHub repo, and which algorithms are next in line.
Read article →Learning With Dr Neal
Professional differentiation. In medicine, law, finance, engineering, and research — every field that processes data — practitioners who can build and evaluate AI tools will command significant advantages over those who can only consume them.
Read article →Taking Claude Fully Agentic
Superclaude = claude --dangerously-skip-permissions. The transition from manual coder to system architect is the biggest productivity jump in my 40+ years of electronics development.
Read article →The AI Reckoning
Understanding the risk posed by AGI is like standing at the foot of an accelerating train track — you can debate the speed or the destination, but wisdom lies in stepping off the tracks before exponential acceleration leaves you no time to move.
Read article →From High School Grad to AI Expert: How I Train State-of-the-Art Deep Learning Practitioners
For many years I have trained more than 600 AI practitioners — from medical doctors to high school graduates — transforming coders into effective AI engineers. Here is the approach that works.
Read article →