I couldn’t speak. It turned out to be a mini-stroke. At 37, I thought they were joking. After a week in the hospital, I started thinking about what we, as a society, have taken for granted. Let’s think about how we approach health. We go for an annual check-up. We test almost the same panel that a twenty-year-old marathon runner and a sixty-five-year-old diabetic would get. The reference range on the lab report also isn’t tailored to your risks. Despite that, they tell you everything is within range and that you’re healthy. You put the results in a drawer and never look at them again. You think you’re being responsible. Same thing next year. It never occurred to you to look up the official recommendations from medical bodies and the scientific community — what to test for your risks, in line with your family history and your stage of life. Why? Because it’s complicated. Scattered across dozens of institutions. Pretty hard to understand. And doctors don’t do this for every patient until something goes wrong. The best you do is ask ChatGPT or Claude and hope it doesn’t make up an answer. I refused to accept that, in the 21st century, this is all we have. So, guided by dozens of the most respected medical bodies, I built a deterministic algorithm that analyzes more than three million clinical data points, extracts official recommendations for your risks, creates a custom panel, and generates a report that often exceeds 30 pages — with personalized ranges for each of the 400+ supported tests. It’s not a replacement for a doctor. It’s support for the healthcare system. So that “he was healthy, I don’t know what happened” is no longer something the neighbors end up saying. Comment “health” if you want me to send you an example of what your next check-up report could look like. P.S. We’re currently preparing for launch. If you have recommendations for journalists who might find this topic interesting, let me know. Labs from 190 countries are supported, with reports available in 20 languages.

Hey, have you ever tried copying text from ChatGPT into Google Docs? It's a mess, right? Formatting goes haywire, and you end up spending more time fixing styling than writing changes. Imagine this instead: