Zach Lipton

And there's places where of like extreme competence.

And there's also there's areas where we're guessing it's like kicking the vending machine.

I think autoimmunity is that kind of situation where we don't really have a very strong mechanistic understanding.

I think it's improving a little bit in recent years, but.

I think as a patient navigating the system, realizing that like you're in this like nebulous cohort of like hundreds of thousands of students that sit somewhere between the diagnostic lines and you're trying to triangulate like what is the state of medicine based on case reports of two patients here, 10 patients there.

And there's a little bit of feeling of like we have to be doing something smarter.

And I tried to put my finger on like what is it that we could be doing more intelligently?

And I think there is this mix of part of the problem was like just people didn't even have access to the right data.

Part of the problem is the data, the right data wasn't even always being captured.

Part of the problem was that even if you had all the data from all the systems in the world, there's still some fundamental mathematical problems of how do you go from longitudinal data?

to actionable, you know, sort of recommendations that would actually improve outcomes for people.

I mean, that's a fundamentally hard problem that we call causal inference.

academia.

But, you know, to be able to produce action guiding insights without, you know, being completely bottlenecked only on, you know, isolated, like randomized control trials, to be able to actually harness the entirety of like longitudinal data to produce those kinds of insights requires a set of strategies that, you know, people in statistics and public health and econometrics have been studying for years.

So for me, I think I came together with this like wild idea in 2012, which was, hey, like, if I want to have an impact on healthcare, like the route to do it would be to do a PhD in machine learning, which, you know, maybe in 2026 seems like extremely obvious, but...

At the time, it was a quaint community.

There was a little workshop that I actually ended up, I'm still deep in that community, but there was a workshop that was called Meaningful Use of Complex Medical Data, which actually gave you the lovely acronym MUCMED.

And this was like, it wasn't even a publishing conference.

It was a group of like,

I don't know, 70 people that would get together like once a year at Children's Hospital Los Angeles and have like poster presentations.