Naji Gehchan: Welcome to SpreadLove in Organizations, the healthcare leadership podcast where we explore leadership with purpose.
I’m Naji, your host, and I’m excited to be joined today by Jin Kim, founder and CEO of Miracle, a software company helping biopharma and medical device companies gain real-time visibility into clinical trials.
Prior to founding Miracle, Jin worked with leading pharmaceutical companies and health systems on data interoperability and analytics initiatives. Outside of work, he previously trained at the Olympic Training Center as a member of the U.S. archery team.
Today, he works with clinical-stage companies to improve how studies are monitored and managed, helping teams reduce manual trackers, identify operational risks earlier, and bring new treatments to patients faster. Jin, it’s great to have you with me today and to see you again.
Jin Kim: Yeah, thanks for having me. It’s been a while.
Naji Gehchan: Before we talk about clinical trial technology and all that you’re doing in this space, I’d love to start with your story. You trained at the Olympic Training Center as a member of the U.S. archery team, which is fascinating. I’d love to hear what that journey taught you about yourself and leadership, and how it has shaped the entrepreneur you are today.
Jin Kim: Yeah, absolutely. Happy to start there. It’s been a while, but having been on the U.S. archery team growing up, I spent a lot of time out on the field with my bow and arrow, aiming at a target 70 meters away. That’s just a little shorter than the length of a football field, so it’s pretty far. A lot of my training involved shooting hundreds of arrows every day.
But one of the aspects that really shaped my mindset was the mental game. Oftentimes, when you’re looking at golf players or archery athletes, everyone at the Olympics, as an example, is capable of shooting a bullseye. But I think the question comes down to: Who can do it on the biggest stage, when the entire world is watching? Can you shoot a perfect, dead-center bullseye right then and there?
We trained so much on the mental game. What that’s really taught me applies to starting a company, working with customers, winning new deals, and trying to do things that no one else has really done before. The entire world is essentially telling you no.
But you still have to believe in yourself. In the early days of Miracle, as we were fundraising, trying to get new customers, bringing on new team members, and growing the team, you get told no so many times.
But I’ve stuck to it. Despite all the rejections, I didn’t let them get to me, and I’ve stayed focused on the vision.
I started Miracle with the vision of helping accelerate drug development, and specifically clinical trials. I stuck to that mission, believed in myself, believed in that mission, and just kept at it. Ultimately, archery definitely played a big role in helping me stay focused and stay on track.
Naji Gehchan: And why this mission? Let’s go from archery and those great lessons you shared to why this mission. What brought you to really focus on this problem that we’re all trying to solve?
Jin Kim: That’s a really good question. Doing archery, I thought it was really hard to shoot the bullseye. But there were other Paralympic athletes who were shooting the bullseye way better than I could, and I saw how they were using prosthetics and medical devices to help them pull back the bow and shoot.
I really fell in love with how these kinds of innovations help people overcome challenges, despite missing a limb or having an abnormality. In fact, many of those athletes can shoot way better than I can.
That’s how I ultimately developed an interest in medicine. Having gone to MIT, I have a technical background, and my initial goal a long time ago was to become a physician like you, Naji.
With that interest in medicine, I wanted to help develop new therapies and treatments that could help people overcome their daily challenges.
Paired with my technical background, that really opened my eyes, especially in the early days when I was shadowing physicians. Every single hospital room has a computer, and there are databases capturing all kinds of patient data.
Having had some experience working with pharmaceutical companies and health systems in a sales role, I learned how many different stakeholders there are, how to align incentives, and how to factor in regulations around data privacy and security.
I realized that I had a passion for medicine, knew how to build software, and now understood how to align and work with different stakeholders. I found myself at a very unique intersection.
Ultimately, with this kind of background, I didn’t know what to do four or five years ago. So I started picking up the phone and reaching out to people in my network. Naji, I reached out to you four or five years ago, in the very early days when I was thinking about what I should even be working on with my background. You were so generous with your time, telling me about your background and what you do. I believe you might have been at Eli Lilly at the time.
Naji Gehchan: Yeah.
Jin Kim: You shared your experiences in Big Pharma. One interview at a time, I ended up doing 150 or 160 interviews that summer.
That’s where some of the biotech companies I was talking with started showing me how, when they run Phase 2 and Phase 3 studies, they have all kinds of spreadsheet trackers.
I knew you could create integrations to bring all that information together and do it automatically. But because I had a keen interest in medicine, I also really cared about the new therapies they were trying to advance.
The short story is that my different backgrounds came together about five years ago, and I had an aha moment: This is what I want to work on. I really want to leverage all the different backgrounds I’ve had.
Ultimately, the goal is to improve healthcare. The way I want to do that right now is by helping accelerate clinical trials as a way to improve the overall drug development process.
Naji Gehchan: Wow, I love what you’ve done. It takes courage to rethink yourself and reach out to so many people. With 150 interviews, you’re getting close to what we have on this podcast. We’re approaching 200 episodes soon!
It’s always great because you gain energy and knowledge, and you ended up building a successful company by understanding the problems. I remember when we first discussed these things.
What was the most frustrating part of clinical trials that compelled you to build a company and a solution to solve it? I certainly see the real impact. By accelerating trials, you’re accelerating innovation for patients and helping hundreds or thousands of people.
What was that one problem in the very beginning that made you think, “Oh my God, this is the most frustrating thing, and I know how to solve it. We’re going to build a company around it”?
Jin Kim: How I found the problem we’re working on right now is an interesting story. This was pre-ChatGPT, back in 2022.
I read some papers about large language models and natural language processing. As I was talking with pharmaceutical companies, they told me how painful it was to redesign protocols and make amendments.
A large part of that work involved not only understanding their clinical trial, but also researching online to see what other publications and studies had been done, what they could learn from them, and what they could take away to drive upcoming protocol amendments.
So I put together a quick AI solution that could identify research papers they should read based on similar studies and recommend how they might improve their clinical trial protocols.
The only thing was, I had a couple of biotech companies that said they were interested, but no one used it.
When I talked with their heads of clinical operations and chief medical officers, they said, “Yes, this is a big problem, and we’re going to use it someday.”
But as I continued talking with them, and some gave me the opportunity to speak with their broader teams, I realized that although they were going through protocol amendments, no one was really talking about whether they’d read a particular paper or how they could redesign a study.
A lot of their focus was on recruitment vendors and their electronic data capture systems. They were trying to figure out how much money to spend on recruitment vendors, how to turn recruitment around, and how to hit their enrollment goals quickly.
In some of those meetings, people were pulling out spreadsheet after spreadsheet, logging into different systems, and copying and pasting information into trackers.
I started asking, “Why are we doing this manually? Is there a way to automate it?”
The short answer was no. But I didn’t take that no for an answer.
With one of the teams, I essentially pulled an all-nighter looking at the systems they were using and the APIs or data access available. I came back the next day with an automated solution for the spreadsheet tracker their clinical operations team was maintaining manually.
I had an automated version of it.
At first, it was a hard sell because it was a different workflow from what people were used to. A lot of the industry is accustomed to saying, “This is how we’ve run studies at our previous companies. It’s worked for us before, so it’s going to work for us here.”
What I was doing was different.
It took about two weeks with this particular biotech team. Their clinical operations team continued creating the spreadsheet trackers, while I also provided my automated solution in the very early stages of Miracle.
The numbers were the same. I was simply automating exactly what they were doing. By the end of those two weeks, people started realizing, “You mean I can spend 30 or 40 minutes, or even hours on some trackers, doing this manually, or I can just go to Miracle and get the automated insight?”
Naji Gehchan: So, fast-forward: you’ve built the company and hired the team. Where are you now? What are you guys doing?
Jin Kim: About three and a half to four years in, it’s exciting that we have our own office in San Francisco. We’ve been able to support dozens of concurrent studies across Phases 1, 2, and 3 in global trials.
We’ve expanded into various therapeutic areas, including Parkinson’s disease, autism, cardiovascular disease, and rare diseases, just to name a few.
But I think the really exciting part is how much we’ve grown. Three or four years ago, we started by helping teams track recruitment: What are recruitment vendors doing? How many people are screening? How many people are enrolling?
We’ve now expanded into a vast number of use cases across the industry. We’re helping monitor patient profiles, labs, safety, drug supply, finances, and enrollment projections.
A lot of these reports come from electronic data capture, IRT, and CTMS systems.
We’ve really focused on understanding what clinical operations and data management teams have to do manually, and what reports aren’t available out of the box from EDC or CTMS systems. A lot of those are cross-system reports.
For example, if your team is tracking drug supply and wants to compare it against upcoming visits, you can project all future anticipated visit windows once you have a baseline visit. That gives you a baseline for how many kits you want at each site.
Today, many teams, if not all of them, are doing this manually. You go into your EDC system to find when patients had screening and randomization visits, then project all the upcoming visits.
You go into your RTSM system to look at all the kits on-site. Then you create a tracker to compare the number of kits on-site by month against the number of upcoming visits.
That’s just one example. The data is already in Miracle. We integrate with EDC, IRT, labs, CTMS, and other systems.
With these kinds of cross-system reports, you don’t have to lift a finger. It’s all there.
We’ve really been focused on helping teams save time. It’s data they already have, and we’d rather have teams spend their hours actually working on the study, not copying and pasting numbers into spreadsheets.
Naji Gehchan: This is great. You’ve built a company around a real problem and a use case for AI and other technologies.
A lot of people talk about these technologies. I think few are doing something meaningful, as you are. You’ve been proving that it works and that we can accelerate things.
What do you think people misunderstand most about how these technologies can improve clinical development and clinical trials specifically?
Jin Kim: That’s a really good question. Naji, you and I go to conferences, and the talk is all about AI everywhere.
I think AI and these kinds of technologies are often treated as a black box. People think, “This can turn our studies around,” or, “This can fix everything.” On the other extreme, because it’s a black box, people say, “We don’t know anything about it, so we need to stay away.”
At least from what I’ve seen, some of the best companies implementing and adopting new technologies like AI start by understanding the process they’re going to use.
Going back to the drug supply example, your team first needs to have a way of doing it, even if it’s manual or slow. Only by understanding the process can you supercharge it with technology later.
Once you have a repeatable way of creating that tracker, it’s easy to bring in an engineer or use an AI tool and say, “This is the exact workflow. This is the expected outcome. Now go and automate this workflow.”
Ultimately, accountability and ownership remain with the team, no matter what technology vendor, API, or AI you use.
The best teams I’ve seen implementing and adopting AI are the ones that take ownership, really understand the process inside out, and recognize that AI and these technologies are simply a means of helping them do things a little faster than they could manually.
Naji Gehchan: Yeah, yeah. What is the first word that comes to mind when I tell you leadership?
Jin Kim: Leadership. The first thing that comes to mind is example: leading by example.
Naji Gehchan: What about clinical trials?
Jin Kim: The first word that pops into my head is spreadsheets.
So many teams we’ve crossed paths with start by showing us all their different trackers and Excel spreadsheets. I think there’s a running joke that this industry runs on spreadsheets.
That’s the first word that comes to mind.
Naji Gehchan: What about AI?
Jin Kim: The first word that comes to mind is ChatGPT, because it’s a tool I love using.
Naji Gehchan: And what do you love about it, especially since you’re developing these kinds of tools?
Jin Kim: To be fair, AI is a small part of what we do. I think the bigger thing we do is provide the foundation.
We see a growing number of biotech and pharmaceutical companies wanting to use AI. But the key thing is that, if you want to use AI, you need data.
We’ve created that foundational piece: a data pipeline from your EDC, IRT, and lab systems into a single database that not only consolidates the data but also normalizes it into a usable format. Then your AI can run on data that’s all in one place, as opposed to running on data silos and unmapped data.
Naji Gehchan: Yeah, that’s often a fundamental problem that no one actually tries to solve. You start adding tools, but, as you said, the real question is whether you have a consolidated database that’s clean and can communicate with the different systems where the information sits. Being able to do that is really critical.
Jin Kim: Yeah, and there’s more and more data coming into studies that we’re capturing about endpoints, procedures, and other things.
There’s just so much data, and I think the only way we can truly make sense of all the different data we’re capturing in a study is by first creating that kind of data pipeline and having a platform on which you can reliably run AI.
Some of the examples I’ve seen at conferences involve teams that are mesmerized by AI tools like Claude and Copilot. I’ve heard of teams that take their EDC exports and upload them directly into Claude because they want people to chat with their data.
There are a number of questions and alarms that go off in my head when I hear that.
One big issue is that if you upload your EDC data into Claude, you’re only going to have visibility into EDC. There’s so much other data in your IRT, CTMS, and all the trackers you have.
You’re only going to get a small piece of the puzzle that way, not to mention the data privacy and security concerns.
Naji Gehchan: Yes. The last phrase I’d love your reaction to is “Spread Love in Organizations.”
Jin Kim: Spread Love in Organizations. The first word that comes to mind is you, Naji.
One of the things I’ve always felt is that I’ve been very grateful, especially as I’ve been running Miracle over the past four years.
I still remember the very first conversation you and I had. I can only imagine how busy you were leading a large team at Eli Lilly, but you still took some time out of your day to help me. I believe I was a graduate student at MIT at the time.
You were so gracious with your time, spending a couple of minutes telling me about your work and clinical trials. You helped me understand the industry, and I’m always grateful for that initial conversation because look at what it turned into.
I think you played a pivotal role in helping me build Miracle, and now we’re able to help dozens of clinical trials around the world. So I really appreciate that.
Naji Gehchan: You’re too generous, Jin. It was always a great conversation. When you reached back out and told me where you were and what you’d built, it was just incredible. Kudos to you. I’m honored to have played even a tiny part in this. Congratulations again on what you’ve done.
Jin Kim: Thank you, Naji.
Naji Gehchan: Any final words of wisdom for leaders in healthcare?
Jin Kim: I think one of the biggest things, especially with AI, is that everyone is beginning to have really easy access to technology. A lot of people are asking, “What else can we do? What else can AI do?”
One thing I’d encourage is to keep thinking about what’s possible, keep pushing beyond what already exists, and keep thinking about what else we can do with these tools.
But at the same time, ground that in the basics. If we didn’t have AI, how would we have done it? Make sure you have those processes in place so you have a way to supercharge and automate them using AI.
Naji Gehchan: This is great advice. Thanks so much again, Jin, for being with me today and for this great chat.
Jin Kim: Thank you so much for having me, Naji.
Naji Gehchan: Thanks for listening to the show! For more episodes, make sure to subscribe to Spreadloveio.com or wherever you listen to your podcasts. Let’s inspire change together and make a positive impact in healthcare, one story at a time.
Follow us on LinkedIn and connect with us on spreadloveio.com. We’re eager to hear your thoughts and feedback. Most importantly, spread love in your organizations and spread the word around you to inspire others and amplify this movement, our world so desperately needs