In 2017, Nishith Khandwala was a Stanford computer science student with an idea—that you could take existing radiology scans, and use AI to check them at scale and detect heart disease and heart attack risk early.
For years, he and classmate David Eng developed the project—but even when it worked, they hit a wall. Hospitals more or less shooed them away, and it seemed like it might be time to move on. But in 2020, Khandwala’s father had a heart attack.
“The cardiologist told us: ‘There was actually a scan in the past that showed he had an increased risk of heart disease, I think we could have caught this earlier,’” said Khandwala, now CEO of Bunkerhill Health. “Can you imagine? Think about the millions of patients for whom you could prevent those kinds of heart attacks. The key thing though was: We weren’t unique in having a great idea. Great ideas are everywhere, we just don’t translate them into reality.”
Bunkerhill—which Khandwala cofounded with Eng in 2019—takes its name from a single-season 2010s CBS TV show that was decisively panned, but had a central, sticky idea: that medicine can iterate faster. The startup, then, deploys AI agents to handle the tasks that hospitals most need to solve, from long wait times to missed follow-ups to paperwork backlogs. Bunkerhill doesn’t suggest use cases so much as hospitals ask if the company’s agents can solve a problem.
Currently, the company works with 15 health systems via its Carebricks platform, including Cleveland Clinic, Mayo Clinic, Ballad Health, Intermountain Health, Sentara Health, Endeavor Health, and The University of Texas Medical Branch (UTMB) Health. Embedded in its platform, the company has both operational agents and nine clinical FDA-cleared AI algorithms, including one that detects silent heart valve disease early and another that assesses osteoporosis risk.
Bunkerhill disclosed its funding for the first time this week exclusively to Fortune: The startup recently closed its $25 million Series B, led by Khosla Ventures, bringing its total funding to $55 million. Sequoia, Felicis, Optum Ventures, and Y Combinator also participated in this latest round. Vinod Khosla—known as an early OpenAI backer and Sun Microsystems legend—has been backing healthcare companies for decades, and he thinks hospitals are more eager to adopt new technology than ever. AI is a mandate now, he said.
“Software used to be a pain for hospitals,” Khosla told Fortune. “There wasn’t a push to adopt it, other than the medical record and it sort of got in the way. Now, every hospital system is trying to adopt AI. When you switch the words ‘AI’ from ‘software,’ everyone wants to and needs to talk about it.”
That’s certainly been true for UTMB, where 22 of Bunkerhill’s agents are deployed. To UTMB chief AI officer Dr. Peter McCaffrey—though the doctor-patient relationship must remain “sacrosanct”—AI is uniquely positioned.
“We may be in an AI bubble overall, but if there’s one place where the need for AI is extremely legitimate, I’d say it’s healthcare,” said McCaffrey. “We don’t need superintelligence to solve our biggest problems. We need average intelligence.”
In recent months, especially, I’ve noticed a minor tidal wave of funding flowing from VCs into healthcare AI admin startups. It’s given me pause. I bring my hesitation to Alfred Lin, partner at Sequoia, who’s been backing the company since leading Bunkerhill’s $6.5 million seed round in 2023.
“If your premise is that there are too many, I think it’s great that there’s too many,” said Lin. “Then, there’s innovation and competition and hopefully the best ones win. If it’s too few, that would be a bad thing. I much prefer having 1,000 flowers bloom, and the best ones will become durable over time.”
Lin—known as Sequoia’s lead backer of DoorDash, Airbnb, Citadel Securities, and Reddit—said that he’s drawn to industries characterized by regulatory capture, and that Bunkerhill fits squarely in that paradigm.
“Like real estate with Airbnb, or Uber and DoorDash, there’s a regulatory capture in those industries, and healthcare isn’t any different,” said Lin. “What [Bunkerhill] is trying to do across parties is hard, but it’s also undeniable that there’s opportunity to improve health outcomes.”
Khandwala has his own take on my question ‘are there too many AI healthcare admin startups?’
“Why should a hospital need to work with 100 different companies to solve 100 different problems?” said Khandwala. “Most companies are still solving one narrow problem, and we’ve moved past that.”
See you tomorrow,
Allie Garfinkle
X: @agarfinks
Email: alexandra.garfinkle@fortune.com
Submit a deal for the Term Sheet newsletter here.
Joey Abrams curated the deals section of today’s newsletter. Subscribe here.
This story was originally featured on Fortune.com