About

I was interviewing for orthopaedic residencies when I decided not to become a surgeon. To understand why, you have to start on a cattle farm in rural Ohio.

The work was physical, the community was small — the kind of place where you grow up working with your hands and looking out for your neighbors.

For as long as I can remember, what I wanted was to be a source of good in the world — to inspire, uplift, and help those who need it, and to become a reason for hope. My mom was a nurse practitioner, and growing up around her work, medicine was the first place I could see how to do that.

Between the farm and college, I spent two years as a missionary in South Korea, walking the streets of Seoul and learning enough Korean to spend my days teaching — and learning — about God. It was about as far from the Appalachian region of Ohio as I could get, and the first time I gave myself fully to something bigger than my own plans. That set a pattern I never really broke.

I came back and walked on to the rugby team at Brigham Young University — no scholarship, no recruiting, just earning a spot — and we won a national championship while I was there. Still one of the things I'm proudest of.

After college I started medical school. The plan was orthopaedic surgery — I liked the directness of it, the mechanical problems with mechanical solutions, the satisfaction of fixing something with your hands. For years, that was the plan.

A few years in, when I was as broke as most medical students are, a pet of mine needed life-saving surgery I couldn't afford. I wasn't going to let the bill be the reason I lost an animal I loved, so I put the surgery on a credit card and started a hardwood-furniture business to pay it off. The building came easy — I'd spent my life doing manual labor on the farm and in timber-frame construction, handy with tools long before I had a shop of my own. After several months (and a lot of black walnut), the garage operation grew into an industrial-grade workshop — the first thing I ever built that was mine.

Those same years, I spent two summers on global-health expeditions, treating patients and assisting in surgeries in sub-Saharan Africa and the Himalayas. This was the medicine I'd wanted to practice all along: caring for people with almost no access to it. But it planted a worry I couldn't talk myself out of — a surgeon can only help the patients who make it to him, and I wasn't sure orthopaedics would ever take me back to the ones who needed help most.

Late in med school, I got quietly hooked on machine learning — nothing to do with what I was training for, just something I kept coming back to. So when a research year came up, I sought out a clinical AI position at Harvard Medical School, mostly to get to the bottom of what had been nagging at me. I taught myself to code to keep up with the engineers around me. About six months in, ChatGPT launched. I hadn't seen it coming, but it confirmed the instinct I'd been following on my own, and what had felt like a private interest suddenly looked like where all of medicine was heading. I didn't know what that meant for me yet; only that the question had stopped being whether AI would change how disease gets treated, and started being who would do that work.

Even then, the plan was still surgery. What changed it was the timing — if AI was going to remake medicine, it would do it over the same thirty years I'd have spent as a surgeon. And it spoke to the worry those summers had planted: a scalpel only ever reaches the patient on the table, but software could reach the ones who'd never get to a surgeon at all.

So I had a choice. I could stay on track for a secure, respected career — more money than a kid from a Rust Belt cattle farm is ever supposed to walk away from — or I could step off and help build it myself.

I was already interviewing for residency when I made the call. When the Match came, I didn't put my name in.

I enrolled at Carnegie Mellon for computer science instead, to learn AI from the ground up — deeply enough to build new systems for medicine rather than apply off-the-shelf tools to clinical data. Along the way I published some theoretical physics too, mostly out of curiosity: a different field, but the same pull toward understanding how things actually work.

Looking back, every chapter has been the same move: teach myself a hard new field from scratch, and build something real in it — hardwood furniture, then code, then a little physics, now the machinery of the cell. A virtual cell is a problem that lives where physics, code, and biology meet, which is to say right where I've already spent my life. It's the one skill I'm sure I have, and the one Galen needs.

Now I'm in San Francisco, building Galen Health. We're working toward a virtual cell — a model of the cell's machinery accurate enough to run experiments on: perturb a gene, simulate a drug, watch what happens in silico instead of at the bench. The point isn't to memorize biology but to capture cause and effect, so it can predict what it was never shown. Get that right, and two things open up. You can study disease from first principles, tracing what's actually broken and what would fix it instead of guessing by trial and error. And you can run a single person's biology through it to find the treatment that fits them — then do the same for millions of people, not just the few who make it into a trial. Cancer is where we start, because it's where we're needed most — where we can do the most good. It'll take years, and the system is far from finished. But it gets a little smarter every day — and that's the part I find most worth doing.