At 2 a.m., in the middle of a bleeding cardiac case, a junior doctor handed him a printout of test results and asked what to do. Ahmed Elsonbaty, a cardiac anesthesiologist, walked out of that operating room asking a different question: what happens when no senior doctor is in the room? He had never built software before. His tools now run in more than 150 countries.

⏱️ Chapters:
0:00 Introduction
0:30 A bleeding patient at 2 a.m. and no one else to ask
1:31 Why there is no time left to teach juniors the old way
2:33 Why he stopped waiting for an engineer to build his idea
3:21 He started with one tool that read one test
4:37 Why he refused to let generative AI answer doctors
5:28 The AI writes the skeleton, the doctor fills it in
6:58 He removed more from the app than he put in
7:57 10,000 doctors in 150 countries and zero formal training
8:39 The nights his family lost him to the build
9:04 What other apps miss about the patient on your table
10:06 The email that told him his tool could be trusted
11:25 Fail, repeat, and do not skip peer review
12:50 Take home messages

About this episode:
Ahmed Elsonbaty is a cardiac anesthesiologist who had never built software until a 2 a.m. case, a bleeding patient, and a junior colleague holding a printout made him ask what happens when no senior doctor is in the room. He describes building a first tool that read a single test, then expanding it to cover ventilation, ECMO, and full anesthesia planning. He explains why he deliberately kept generative AI out of the answers, using it only to draft the code skeleton while he filled it with published, traceable sources, so the same question returns the same answer every time. He talks about removing more from the application than he added, testing every addition against whether a doctor could actually use it mid-operation, and rebuilding around suggestions that arrived from strangers around the world. He is candid about the cost: months of nights away from his family, repeated failures, and no formal training to fall back on. When he could not tell whether his own work was any good, he submitted it to the leading journal in cardiothoracic and vascular anesthesia and got an acceptance email from Joel Kaplan, whose name was printed on the textbooks he studied from. The tools now reach more than 10,000 doctors in more than 150 countries, and he insists throughout that they help a physician decide rather than replace one. His closing advice is practical: name the problem precisely, build on sources you can defend, expect to fail repeatedly, and put the thing through peer review before you ask other doctors to trust it.

🤝 Partner with me on the KevinMD platform:
With over three million monthly readers and half a million social media followers, I give you direct access to the doctors and patients who matter most. Let's work together to tell your story.

➡️ PARTNER WITH KEVINMD: https://kevinmd.com/influencer
➡️ SUBSCRIBE TO THE PODCAST: https://www.kevinmd.com/podcast
➡️ RECOMMENDED BY KEVINMD: https://www.kevinmd.com/recommended

#MedicalAI #Anesthesiology #PhysicianInnovation