Dr. Kulkarni talks with Krystal Lin, MD, co-founder of Mozibox, about the problem that pulled him into Predoc, how the team landed its first customers, and why he believes physicians have to cut back clinically before they can build something real.
You trained in ophthalmology and then neuro-ophthalmology. What problem in your own practice led you to Predoc?
I did the traditional path: med school, residency, then fellowship in neuro-ophthalmology. It's an outpatient subspecialty, and most of what I see are second, third, and fourth opinions. By the time patients reach me, they've often had multiple prior consults, labs, MRIs, and pathology.
I had the same problem whether I was at Mount Sinai on Epic with Care Everywhere, or in a suburban outpatient office not connected to any hospital system. I'd walk into the exam room, the patient would be there for help, and I'd have maybe 0 to 50% of their records in front of me. The records were supposed to be faxed over, or sent by the referring physician, or available in Care Everywhere or the HIEs, but they just weren't there. This was incredibly frustrating for me and, more importantly, for the patient.
Everyone loves to talk about interoperability, TEFCA, QHINs, HIEs, and Care Everywhere, and from the outside it looks like a solved problem. I thought it was too. But when I started to dig in and asked why I was the only one who didn't have the records, I found out it's actually not solved, and it wasn't just me.
Predoc was recently featured in Christina Farr's Second Opinion with a contrarian take: the fax machine isn't going away. Why do you believe that?
One of my cofounders is a nephrologist at Morristown Memorial in New Jersey, and interns, residents, and even attendings there still spend a good chunk of their time by the fax machine trying to get charts. Another one of my cofounders has a finance background, and his reaction was: I have all of the world's financial information at the push of a button on a Bloomberg terminal, so why doesn't that exist in healthcare?
When you really dig in, there are a lot of misaligned incentives. Our core thesis is that there's never going to be a magic API in healthcare like there is in finance. You can get maybe 50 to 75% of the way there on a good day, but there will always be a last mile: a long tail of records sitting in the hundreds of EHRs besides Epic, siloed and only available by fax or snail mail. That's the part we automate at Predoc.
How did you go from noticing the problem to joining the company?
My previous startup got acquired and I was thinking about what to work on next. I made a Notion document listing all my pain points as a physician, and medical records kept bubbling up to the top. I talked to maybe 70 to 100 people about different startup ideas, and this problem kept coming up.
One person saw the problem too, and we workshopped it a bit, but then we both got busy. About three weeks later he texted me that I should meet a team in New York that was working on it, and he made the connection. We became friends, they actually gave me a login to the platform at dinner one night, and when I started using the platform in my own practice, that's when the light bulb went off. Once it's automated, you can't go back to sending your staff out of the room to phone, fax, and chase charts. I was so excited about solving the problem that I wrote a very small angel check into their seed round, and later joined as a cofounder. I didn't think too much about it at the time. It just felt like these are my friends and we were working on a problem that everybody has, but nobody was thinking about.
The team already had a physician co-founder. What did you bring?
She and I have complementary skill sets, but we both understood the problem deeply, on a visceral level. She built the whole initial operations side of the business: actually calling offices, faxing HIPAA authorization forms, and retrieving these records from siloed, hard-to-reach locations.
My passion was more on the go-to-market. My previous company was an eye supplement company, and a supplement company is essentially a marketing company. There's that old adage that it's much easier to sell a painkiller than a supplement, and it's true, but the flip side is that some of the world's best marketers come from the supplement space for that very reason. Mine was a very unique eye supplement that I truly believed in and put my name behind as a doctor. Patients and customers loved it once they discovered it, but it's still an extremely tough and competitive industry. I had to go very deep on Facebook ads, building websites, CTAs, funnels, copywriting, storytelling, cold outreach, objection handling, testing different messaging angles, understanding human psychology, building video sales letters, the whole works.
Most physicians have an aversion to sales and are really bad at it, especially when it comes to selling themselves, but I think that's a huge mistake. I'll always be a physician at the core, but I developed a love and appreciation for sales and marketing. I just love the game, especially when you truly believe in your product. So early on at Predoc, I wanted to see if I could help with go-to-market, and some of our initial traction came from LinkedIn and that type of work.
Finding the first customer is where many physician founders get stuck. How did Predoc land its first ones?
Our very initial customers came from our personal networks and some of that work we did on LinkedIn. I think having two physicians on the founding team helped a lot with credibility early on.
The problem of medical records is also so painful that when you describe it, people say, Yes, that is a problem, let me get on this call.
Some of them were hiring people to do it manually.
The other thing that helped was the team. There was an actual product and a whole team behind it. One thing I see a lot of doctors doing is trying to do everything by themselves, and that's where it doesn't work. I made that mistake in my previous company, where I was a solo founder. Here, we have an incredible team that does most of the work, and I can just focus on my expertise.
Do you still see patients? Is it possible to build a startup and keep practicing?
I'm maybe 30% clinical right now. I cut down slowly, and I don't think I'd go too far below this, otherwise it gets hard to have continuity of care for my patients. There's also a lot of value in being clinically active and close to the problem, which really helps me in my role at Predoc. There's not a single day that I'm in the clinic and don't lose my mind over missing medical records.
But one of the hard parts is that you really do have to cut down from full time to have the bandwidth to think clearly about a startup. After a long day of clinic, five days a week, it's very difficult to have the energy to make anything work outside of that. I don't know any doctor that finishes a full day of seeing patients and thinks, That was easy, I have a ton of energy.
We're all drained. So my advice to other doctors is to cut back to at least 80% or so.
When did you make that decision for yourself?
I had already cut back for my previous startup, about five years before Predoc. I had tried it the other way around: working full time and asking myself what my startup ideas were. There was no clarity. I had to cut back first and take the plunge. By the time Predoc came along, I was already down to a small amount of clinic.
Startup progress is rarely a straight line. What were Predoc's pivotal moments?
I agree, it's not linear. What I'd say is that it never feels like you've made it. You just get a new set of problems on the table with each stage, and that never ends. Everyone would love to be the world's most valuable company, but I'm sure if you talk to Jensen Huang, he goes to bed every night with major problems.
For us, the pivotal moments that stand out are getting the first customers and, the really big one, getting the seed investment. Our first institutional round felt like a very big milestone.
Did an accelerator play a role?
Yes. My co-founders were in ERA, a tech accelerator in New York, and it helped a lot. Accelerator terms have also gotten much better for founders over the last couple of years. I would definitely recommend it for any first-time founders.
Beyond physician practices, where else could this technology help? What's next for Predoc?
Hospitals, health systems, and contributing to the AI frontier are next for Predoc. It turns out that the problem of exchanging and organizing records isn't just a problem at physician practices. It's a huge problem for every hospital and health system in the country. Even the big ones that are on Epic with Care Everywhere have huge internal or external manual teams doing it.
There have also been a lot of regulatory tailwinds in the industry, with the CMS Health Tech Ecosystem and a movement toward making it easier for patients to access and share their own records. More and more consumers are actually requesting, accessing, and inputting their own records via ChatGPT and personal AI agents. I think patients using AI to understand their personal health is one of the main use cases for AI, and I think it's great. The frontier labs are also very interested in the world of medical records, because it can help with drug discovery, rare diseases, and all of life sciences. Predoc sits in the middle of all of this, so I'm super excited.
