MFN recently spoke with Eliot Paynter, founder of Tendral Health, whose AI systems identify, verify, and recruit physicians for paid research studies, replacing the static clinician panels the industry has relied on for twenty years. This spring, Tendral Health won MFN’s inaugural Applied AI B2B Challenge program and received a $20,000 innovation grant. Tendral is still a team of one, and on track to clear $300,000 in its first year in production.

MFN: Please give us an overview of Tendral Health.

Eliot Paynter: Healthcare companies, ranging from medical education to pharma, constantly need input from physicians to validate clinical content, inform drug development, and review protocols. Almost all of them rely on the same tool, the clinician panel: a static database of physicians who opted in at some point, sometimes years ago. That model hasn’t changed in twenty years.

There are three main problems with this. First, you can’t typically target below broad specialty buckets. Second, a large share of the budget goes to physicians who turn out not to qualify once they’re already in the study, so that money is simply wasted. And third, there’s often no reliable proof that the person who completed a study is the credentialed physician they claimed to be. At Tendral, we replace the clinician panel model with AI.

We hold a profile for effectively every individually licensed clinician in the U.S., approximately seven million people, built from public federal registries and licensed clinical datasets. Two of our production models sit on top of that: one that scores whether a clinician is genuinely right for a study, and one that scores whether they’ll actually respond. Language models run the workflow around them, from reading a client’s brief to building the study.

The physician side matters just as much. Most research outreach is a mass mailing, and the reward for answering is often a screener that disqualifies you ten minutes in. We’d rather send far fewer invitations and have every one actually fit, so nobody gives up their time only to be screened out.

MFN: You recently relocated from the West Coast to the Boston area. What factored into your decision to relocate here? How does Massachusetts compare to California’s startup ecosystem?

Eliot: Boston is one of the highest-density markets in the world for life sciences, healthcare innovation, and pharmaceutical decision-makers. In this industry, in-person relationships still carry real weight, and being near the people who buy what we build changes the pace of everything.

I also went to school on the East Coast. Having grown up in the San Francisco Bay Area, I wanted to open my horizons and get out of my comfort zone. Silicon Valley is genuinely extraordinary for AI startups. I just don’t believe it’s the only place you can build one, and Massachusetts is deliberately investing in AI right now, particularly where it intersects with healthcare.

I’ll admit I expected Boston to be all traditional finance and little startup energy. That’s the reputation, and it isn’t what I found. In California it’s ambient; here it’s a network you build on purpose.

Having worked with MFN through the Applied AI B2B Challenge program, I was introduced to a few people who brought me into the regional AI events community.

MFN: Tendral Health won MFN’s Applied AI B2B Challenge this spring. What were some of the lessons you took away from the program, and how did the $20,000 innovation grant help advance your business?

Eliot: It’s worth saying upfront that the prize money has allowed me to move considerably faster than I would have just on revenue alone. However, the biggest takeaway wasn’t the money. I think it was being able to explain my business in simple terms to people outside of healthcare and the AI sector. Working alone, you don’t get that kind of pressure applied to your assumptions. We came into the program having run one pilot study, back in late December and January. We’ve since launched fifteen client studies.

Most of the grant went straight into the platform: agent architectures and automation I could never have justified out of operating revenue, several of which are now core to how it runs. Twenty thousand dollars isn’t much in this industry, but for one person running an AI-first company, it’s a lot of experiments. The rest went to design, which mattered more than I expected. Clinicians are cautious and need to see a credible brand before they’ll give you their time, and when you’re one person and not a designer, that’s the first thing you defer. The new brand and rebuilt product start rolling out over the next couple of months.

MFN: What are the next steps for your company in the coming months?

Eliot: The next few months are mostly about automation: getting more of a study to run itself, from the client’s brief through fielding, so adding a client doesn’t mean adding my time. That’s what makes the rest possible. Expanding the client base is the number one priority, and most of what we’ve run so far has been in hematology and oncology, so broadening beyond that is where nearly all my time goes. Every study also makes the next one better, because the network carries forward and the models sharpen on real response data. The model it replaces starts from zero every time.

MFN: Any advice you’d like to share with other first-time founders?

Eliot: Be tenacious. What usually stops you isn’t wanting to quit, it’s not knowing what to do next, so you wait for it to get clear. Waiting doesn’t tell you what to do next. Doing something does. Treat everything as an experiment and design each one to produce validated learning. Almost nothing I believed a year ago survived contact with real clients, and the business is better for it.

Check out what’s happening throughout Massachusetts by visiting the MFN Events Calendar! Hosting a networking event or specialized workshop? Contact the MFN Team and we’ll help share your event for other founders to learn more!