Patanjali Chary, an early student of Geoffrey Hinton at the University of Toronto, joins Sunny Ray to discuss Fourth Vital, an AI-native upstream kidney risk intelligence platform. After losing his mother to kidney disease, Chary set out to detect physiological decline long before conventional labs, symptoms, or crisis point. Fourth Vital pairs non-invasive edge biosensing hardware with longitudinal AI, building on decades-old physiological research that never reached patients or physicians because it stayed trapped in research labs. Chary describes his path from a computer science and AI degree under Hinton, through twelve years in enterprise software at Oracle and Microsoft, to dual MBAs at Berkeley Haas and Columbia, and finally to founding Fourth Vital. He details pivoting from a device-first company to an AI and data-first one, launching IRB-approved data collection with a kidney unit in Ontario, and choosing to focus first on stage five dialysis patients before moving upstream toward earlier stages. Chary frames kidney disease as part of a 1.6 trillion dollar cardio-kidney-metabolic disease burden, with roughly a billion people affected worldwide and ninety percent undiagnosed until it is too late.
A Geoffrey Hinton student turned founder builds AI biosensing to catch kidney disease years before labs can, tackling a 1.6 trillion dollar crisis.
What does Fourth Vital do, and why should people care?
Close to a billion people have kidney disease, and ninety percent don't know until it's late stage because there are no early symptoms and routine labs catch it too late. After losing my mom to this, I found research on physiological signals that detect kidney decline earlier. Fourth Vital combines an edge biosensing data capture layer with unique AI to diagnose and stratify patients much earlier, giving physicians a real chance to intervene.
What was it like learning directly from Geoffrey Hinton, and what idea from him still shapes how you build today?
I did my undergrad in computer science and AI at the University of Toronto, studying under Professor Hinton before anyone cared about AI. He always believed healthcare was AI's true promise and biggest opportunity for impact. I feel blessed to have been in those early conversations. Ironically, my mom was skeptical of my going into AI, and now I'm using it to detect the disease that took her life.
What took you from twelve years in enterprise software at companies like Oracle and Microsoft into kidney disease specifically?
Losing my mom was the motivation. It became a passion project, and I spent years studying the causes of kidney decline and how it could be detected. My background is in building large, complex enterprise systems that handle huge amounts of data, and my team has that same experience. We're applying those same principles to kidney disease, bringing everything together in a simple interface physicians and eventually patients can use.
How did your two MBAs, from Berkeley Haas and Columbia, shape how you think about building a company in a regulated space like health tech?
I actually ended up with two MBAs, a limited edition dual program between Berkeley Haas and Columbia, with exams on both coasts. I hadn't studied finance before, and I learned how to bring something to market at scale. Many classmates are now involved with Fourth Vital, including on my board. Toronto gave me an interdisciplinary foundation, and business school gave me the tools to execute on it.
What was actually missing before Fourth Vital, given the research on physiological fluid shifts and kidney risk goes back over 35 years?
The research stretches back over 35 years, and I even talked to some of the original paper authors to understand why it never went mainstream. What I found is their solution works in research labs but is basically unusable for physicians and patients. Nobody had looked at it from an engineering and design perspective. Commercializing that gap, making it easy for physicians to adopt and patients to use, was the missing piece.
You now have live IRB approved data collection running with real patients at a kidney unit in Ontario. What had to be true, technically and organizationally, before a clinical site would let you in the door?
It was surprisingly much easier in Canada than the US would have been. The research unit we partnered with was motivated because they already saw value in our software, and the hospital helped us get through the IRB without the complexity I'd have faced elsewhere. It's a testament to a system genuinely motivated to reduce patient suffering and cost. I hope to replicate that in other countries and eventually the US.
You've put the cardio-kidney-metabolic disease category at around 1.6 trillion dollars a year globally. How does a company your size decide where to focus first inside something that large?
We pivoted a couple of times before realizing that stratifying existing kidney patients earlier is where we can get Fourth Vital adopted, especially with value based care incentives in the US. We're starting with stage five dialysis patients because our signal helps with their complex fluid issues, then working backward toward stage one. You have to go where the data tells you the problem is, not guess at a percentage.
Building something daring? Sunny talks to founders like this every day. Fifteen minutes to see if your story belongs on the stage.
Claim your pre-interview