How TJM Labs Reached $25M ARR and 450 Pharmacies with 40% EBITDA (Interview with Jonathan Adly)

What TJM Labs Sells to Pharmacies

Nathan Latka

00:00What is TJM Labs selling to these pharmacies? What what do they get when they pay you?

Jonathan Adly

00:04Yeah. So they get an AI agent. And the AI agent does what a pharmacy technician or a pharmacy assistant does. Pharmacy is an incredibly difficult space. An average retail mom and pop, their margin is 2%. Let's say a grocery chain, their pharmacies lose money. It's a loss leader. You come in and implement is no longer a loss leader.

Nathan Latka

00:26You give us the revenue growth from December 24 to December 25 to today?

Jonathan Adly

00:30Yeah. So 1,000,000 to about 15,000,000 to 25,000,000. We had 40% EBITDA.

Nathan Latka

00:38What?

Jonathan Adly

00:39Yes.

Nathan Latka

00:40That's insane.

Host Introduction and Jonathan's Background

Nathan Latka

00:43Hey, folks. Today, we have a triple threat. Jonathan Adly spent fifteen years working inside pharmacies before founding TJM Labs and combines three disciplines. He's a licensed pharmacist, an AI software engineer, and an operator. He previously worked on AI infrastructure at CaryRx and founded Galen AI. He holds a PharmD, a postdoc residency, and an MBA. TJM Labs is a pharmacy AI platform that automates operational work, prescription intake, data entry, refills, and more. Jonathan, you ready to take us

01:14to the top?

Jonathan Adly

01:15Yeah, I'm ready.

From Pharmacist to Entrepreneur

Nathan Latka

01:16Alright. So take us into the story here first. What came first? Were you running? Did you get your doctorate first? Were you a pharmacist first? Were you a coder first? What started this?

Jonathan Adly

01:27You know, I don't know how far back you wanna go, but, I remember being in high school and we're talking about colleges with my parents and what you wanna do. And, classic immigrant story, they always want you to be a doctor and I did not want to be a doctor. Actually, healthcare in general was not something I'm interested about. So we ended up basically fighting for six months and pharmacy was the compromise. And the idea is

01:52go to pharmacy school, get your license, right, that puts a really hard floor in your career, because if you have a pharmacy license, always have a job. Typically in pharmacy school when you do well, they push you into academia. So I ended up publishing a paper, doing the residency, did some postdoctoral residency work. And I ended up, I would say in a great job, very comfortable in a health system. Eventually promoted to do some admin

02:21work, understand where the money is as far as the pharmacy supply chain, all the government interventions that could be well meaning, but ends up having a secondary side effect. I've seen, I would say how the sausage gets made.

02:39That made me wanna be an entrepreneur. There's a great opportunity there and that's when I started pursuing my MBA. So tried a couple of things, didn't really work out. I was a consultant, I ended up working,

02:52I would say first as a product person because I had the domain expertise, but eventually took on more and more actual engineering tasks and self taught myself. And at some point, you know, I was consulting the same problem showed up again and again and again. And it screamed, this is a product, not a consulting engagement.

Founding and Exit of Galen AI

Jonathan Adly

03:17That's how Galen AI was born. Galen AI did super well and in about a year and a half it got acquired.

Nathan Latka

03:26Jonathan, what does super well mean? Did you raise capital for that company?

Jonathan Adly

03:29I did not raise capital. It was bootstrapped, right?

03:35When we got acquired, we had our first enterprise, like true enterprise customer. And if you have one enterprise customer, usually that means you can get the second and the third and the fourth.

Nathan Latka

03:44And was that a meaningful financial exit for you or no?

Jonathan Adly

03:46It was a 100% a meaningful financial exit. I would say, so Galen AI ended up, it was a very similar software stack, like a company like Open Evidence. Open Evidence, I think it's a unicorn now, like a multi billion dollar valuation. The difference is Galen AI was very B2B focused versus Open Evidence when they're just physicians can sign up and work out of it. Galen AI exit was too early, Everybody would have said you sold way too

04:15early. You should have stuck around.

Nathan Latka

04:17What was revenue when you sold?

Jonathan Adly

04:20I actually can't say because this is like under the NDA when we got acquired. But I can tell you that we first, when you sign an enterprise customer, usually there's like an implementation, it takes some time to ramp up the revenue. It would have been enough again, as a bootstrap company where I can just have this one enterprise customer and not do anything else.

Nathan Latka

04:40Are you talking like one customer paying 7 figures above 1,000,000 a year or was it more like, you know, a 100,000 or 50,000?

Jonathan Adly

04:47No, in healthcare, typically enterprise is more than $250K a year. 500,000 is reasonable, but there is million, 2,000,000 contracts.

Nathan Latka

04:58And then how did you go from Galen AI to CaryRx? Was CaryRx the acquirer of Galen?

Jonathan Adly

05:03CaryRx was the acquirer. We shared, I would say an enterprise customer and it was an easy conversation. They were a great software company. AI was just starting to bloom so to speak. And Galen AI had, I would say the AI stack

05:23very early ahead of the curve. It was a great acquisition for both parties and CaryRx exited six months ago. So they also got they did well and they sold to someone else. And yeah, it was successful for me as a founder to have that first exit, because that tells you you can build at least the zero to one stage. And people don't realize this, zero to one is really hard. You know, one to 100, there's playbooks and

05:47I think can be executed on somewhat programmatically, but zero to one is that's the part when there's no playbook for. You have to go and figure it out.

Why Jonathan Left CaryRx to Build TJM Labs

Nathan Latka

05:57Why did you feel like you had to leave, CaryRx to build this business versus building it inside of CaryRx?

Jonathan Adly

06:03I think I just missed being a founder. Right? I had the vision, especially early in AI, that is gonna evolve extremely fast, that we should not... To treat AI not as software, not like software plus plus, but as true labor replacement. And that was way too early to convince the whole mature and successful organization that that's the way to do it, right? Everybody in that period was thinking like, I want to be the system of record

06:35for this specific slice of industry. And that would be the win. Versus I was like, I don't care about being the system of record. I want to be the system of action.

The AI Agent Product and How It Works

Nathan Latka

06:44But if we fast forward to today here in middle of twenty twenty six, what is TJM or TJM Labs selling to these pharmacies? What do they get when they pay you?

Jonathan Adly

06:54Yeah, so they get an AI agent and the AI agent does what a pharmacy technician or a pharmacy assistant does. So if you imagine, if you walk into a pharmacy, typically the doctor have sent an electronic prescription, which is basically, you know, a regulated version of an email or a fax, or they gave you a hard copy of a prescription that you walk into the pharmacy. Typically, you know, pharmacy has staff that spending the whole day typing

07:23these prescription into their system of record. And then the pharmacist double checks it and then dispenses it to you as a patient. What these bots do is they replace the data entry. And typically one of these agents, bots, is two to three pharmacy technicians.

Pricing: $45K Per Bot Per Year

Nathan Latka

07:40How do you structure your price and what's the average pharmacy paying you today per year to use your technology?

Jonathan Adly

07:47Yeah, so our average ACV is about $45K a year, right? And if you know what it is, that is slightly less than the salary of a pharmacy technician, and that's how we price it. You can hire a digital worker for 45 k a year, or you can hire three human worker, and that will cost you 150 k a year, right, or more if you count benefits and payroll and all of that stuff. An idea is

08:14these agents work twenty four seven. You train them once and they work with you forever. You can always retrain them and make them do additional things.

Nathan Latka

08:22And did you always price this way, even going back to 2020, even last year when you launched?

Jonathan Adly

08:26Yeah, that was the vision. The vision was always digital workers. We didn't think per seat pricing makes any sense in the AI world, right? We don't think task also makes sense. This is very close to tokens.

08:41Tasks could be very ambiguous. Imagine you're hiring someone and you told them you're gonna only work in this one task and that's it. Typically that's not how people work.

First Line of Code and First Customer

Nathan Latka

08:49According to your blog, you officially launched in 2025, but when did you write the first line of code for the platform?

Jonathan Adly

08:55Yeah, we wrote the first code actually November 2024. I remember it was a holiday break, we had our first customer essentially win a really large contract, they had to hire about 200 technicians for and they had four weeks to do that. That is a monumental difficult task to hire this amount of people and train them. So they were pretty desperate and you know, desperate times call for desperate solutions. So that's how our company was born. We

09:26said, let me try it. Let me see how many bots, agents we can put in. So instead of hiring 200, maybe have to hire only 100. Right. But it ended up being extremely successful. And over time, as the AI capability got better and higher, the solutions just kept getting better and the tasks that it can do kept expanding as well.

Nathan Latka

09:50And so that was your first customer, I assume your ability to get that deal was just from your network being at CaryRx and your experience in the pharmaceutical space, correct?

Jonathan Adly

09:58100%. And this is, if you think about it, it was like an immaculate inception because pharmacy is a very regulated flow where there's no test sandboxes, you actually have to work on real prescriptions, real medication that can cause serious harm. So you cannot just have a landing page, I'm going to do this and people will just say, here's the keys to the kingdom because you might actually kill somebody. You have to have a ton of trust

10:25built into the system already over many, many, many years, right? In order for people to say, great, go ahead. Here's access to our production system. Let's go live.

Getting the First 10 Customers Through Trust

Nathan Latka

10:37So how did you convert your years of experience in this space, your network into hard paying customers at TJM Labs? How did you get the first 10 customers?

Jonathan Adly

10:48Yeah. It's all trust. You know? I think the first customer was obviously a relationship that was built over many years. The second customer was literally a friend of the first customer, and he's like, I see what you have done for them. Please do it for us too. The third customer, this is when I got convinced I need a sales team, was one of my customers in an airplane sitting next to a guy and they were chatting

11:17and he showed him like what these agents are doing and the guy came down from the airplane and called my cell phone and said I want that too. And we stayed this way until about 30 pharmacies or so.

Nathan Latka

11:29Fast forward to today, how many customers do you work with today?

Jonathan Adly

11:32We have about 450 pharmacies.

Nathan Latka

11:34The average you gave me earlier, is that per location or per pharmacy?

Jonathan Adly

11:39That's per bot, per digital worker, right? And a pharmacy can get a 100 digital workers if they're big enough.

Nathan Latka

11:45But if you look at your pharmacies today, I imagine some of them have a 100 locations, some of them are mom and pops with one or two locations. What's the average number of bots per your 450 pharmacies would you guess?

Jonathan Adly

11:56I would say probably 2.5, almost three. So...

Nathan Latka

12:01Okay. Okay. Interesting. Can I take that math? I mean, 2.5 bots at $45,000 per year per bot means each of the 450 pharmacies would be somewhere around a $100,000, 112,000 paying you per year. Is that accurate?

Jonathan Adly

12:16Yeah, and then obviously there's segments.

Revenue Growth: $1M to $15M to $25M ARR

Nathan Latka

12:19So we can obviously back into your revenue that way, right? I mean,

Jonathan Adly

12:21Yeah, I mean, I can share my revenue, my revenue is about, it's close to 25,000,000 ARR, right? That's not a secret.

Nathan Latka

12:28And what growth rate does that represent? If you go back to December 2025, where did you finish there?

Jonathan Adly

12:33The first customer I just told you about, we stayed nine months for them. And that's obviously a massive customer, a million plus a year. We only started selling to other pharmacies February 2025. So it's been a year and like five months now.

Nathan Latka

12:50Can you give us the revenue growth from December 24 to December 25 to today?

Jonathan Adly

12:54Yeah. So 1,000,000 to about 15,000,000 to 25,000,000.

Nathan Latka

13:01That's incredible. So the last six months, you've added about $10,000,000 of ARR. How the heck did you do that expansion or new customers?

Jonathan Adly

13:08Yeah, mean, look, it's a good product. We know what we're doing. It's what we call it is a fifteen year overnight success. Pharmacy is an incredibly difficult space. An average retail mom and pop, their margin is 2%. When you come in and implement solution for this, go from a 2% margin to an 8% margin. Who would not want this, right, as a small and bop business in America? These enterprise customers, like I say, a grocery chain,

13:34their pharmacies lose money, it's a loss leader. You come in and implement, it's no longer a loss leader. We had an incredibly strong team, we made two acquisitions during that time,

13:46And yeah, there's a reason we raised the 100,000,000 in like six months, so.

Nathan Latka

13:50How much of the revenue growth was inorganic? It came from the acquired companies.

Jonathan Adly

13:55Less than 5%. We essentially had the acquisition more for the team than anything.

Raising $100M Without a Pitch Deck

Nathan Latka

14:02I see. Okay, let's jump into that story. So you've been pretty aggressive on the capital front. This press release you put out said a 10,000,000 series A, a 15,000,000 A1 and 75,000,000 from Elephant. What was the timing on those?

Jonathan Adly

14:16Yeah. So series A December 2025. Right? Series A1 March, series B June.

Nathan Latka

14:28Take me back to the 10,000,000 Series A in December because you were already doing significant revenues at this point. You said 15,000,000 of ARR. Why raise it? I I guess why raise the money at all? And if not, why not go all in? Why not raise as much as you possibly could?

Jonathan Adly

14:43That's a great question. I didn't want to raise actually. Didn't, to this day we still don't have an investment deck.

14:50The only reason I raised that Series A because there's exceptional, there was two exceptional teams out there, right? They were doing, I would say, adjacencies that very easily can overlap what we're doing. They're both were pharmacies. And I was like, if we can combine, it will be almost like the Manhattan Project. We actually internally called it the Manhattan Project. You combine all of these great minds in one room, there's no one else, maybe like five guys

15:19in the whole world that can do that. We have all five.

Nathan Latka

15:22So just to be clear, you bought... You used some of that 10,000,000 to buy EncoreRx and Pharmesol, or was the original 10,000,000 just for EncoreRx?

Jonathan Adly

15:3010,000,000 was EncoreRx, and then 15,000,000 was for Pharmesol. Obviously, the acquisition numbers are not exactly the same, and we did some other stuff with the money, not exclusively acquisitions.

Nathan Latka

15:40And the combined revenue of EncoreRx and Pharmesol, all you said was under, like, 2 or $3,000,000 when you bought them, quote 5% of your total ARR?

Jonathan Adly

15:48Yeah. Yeah. So... Yeah.

The Elephant Series B and Secondary Liquidity

Nathan Latka

15:49Okay. Okay. Really team led. Okay. Take me into the Elephant Round series b, which just closed couple days ago. Why do this deal and was this all primary or secondary as well?

Jonathan Adly

15:58We decided in the beginning, we decided not to announce. So again, we had this mindset of like, we are... There's this company called Veeva, a public company, they're S and P 500. No one ever heard of them. But if you are a pharma company, they are your software provider. Like, our role model is Veeva. How we raise money, how we work is Veeva. So it's the same mentality we don't want to announce. However, we have decided

16:26that announcing is a good idea because as we're hiring and we're ramping up, it helps a lot with talent. People don't want to jump in a company, and then essentially the company goes away. When you say we raise 75,000,000, it just makes conversations a lot easier.

Nathan Latka

16:40Okay, that didn't answer my question though. You also have to manage dilution as a founder, right? So does Elephant come in and put all this on the balance sheet's operating capital or was there some liquidity in secondary for early employees? How did you view the money?

Jonathan Adly

16:52Yeah, there was some liquidity, it was all secondary. I also believe if you raise, there has to be a secondary. I don't believe in raising without secondaries. A lot or a little depends on the dynamic and what you feel the market looks like. We want to have a balanced balance sheet where we never have to raise again if we don't want to, and we're still profitable and continue to be profitable. And M and A, we really

17:14believe

17:16M and A is important. Right now, it's the same as a transition between cloud and on premise, and you have to run, you cannot walk, because that transition is going to finish in like a year from now, and whoever is the AI agent category winner is the category winner.

40% EBITDA and Raising When You Don't Need To

Nathan Latka

17:36Jonathan, you said you were profitable, so you had a lot of leverage going into the Elephant Round. You probably didn't have to do the deal. Can you share? Are we talking like 10% profit margin or more?

Jonathan Adly

17:44No. We had 40% EBITDA.

Nathan Latka

17:47What?

Jonathan Adly

17:48Yes.

Nathan Latka

17:49That's insane.

Jonathan Adly

17:51Yes.

Nathan Latka

17:51Wow. So what were you do... Now I'm even more curious why take external money and take all the dilution. It's just really because you wanna signal to the market, hey. Listen. Come work for us. We've got a lot of money. We want the best talent.

Jonathan Adly

18:02Yeah. Look, at the end of the day, I came from an acquisition already. I sold a company. Right? If I'm strictly thinking about this, like money, like I don't have to work. I can just sit at home and do nothing. The idea is

Nathan Latka

18:17You're a builder, you would never do that.

Jonathan Adly

18:18Yes, exactly. And I tried doing that and it didn't work out. The

18:23idea is you wanna build a company of a significant proportion. You you play want, the game for the love of the game at the end of the day, right? And if you want to win the category, right? If you want to be a classic vertical software company, if you want to beat Veeva, right? If Veeva is for pharma, TJM Labs is for pharmacy, You have to raise money and you have to be aggressive. Right? And you

18:46have to basically run, run, run as fast as you can. And raising, people don't realize this. Every raise is a miracle, because someone actually puts their hand in their pocket and gives you lots of money to go build your company. You don't want to raise when you need to. That's actually the opposite. You want to raise when you don't need to because that's when the easiest way to do it.

Valuation Framework and Path to $100M ARR

Nathan Latka

19:08The biggest question I get is, Nathan, I'm considering raising what valuations are the markets giving AI companies today? I'm gonna assume you're not able to show your exact valuation, but can you maybe give us a range? You know, was it 20 to 30 x or more or less? Coach us up there.

Jonathan Adly

19:22People always sometimes raise on vibes. I always like to look at the public companies and see, right, what is their multiple? And if someone has a 10x multiple, like Figma, when they went public, they had like a 15x multiple or something like that, or maybe 10x multiple. And they were growing 40% a year. If I'm growing 10x, right, then it's not going to be less than 15x a year. That doesn't just make any sense. So even

19:49in these conversations, when you want a valuation, you always want to base on real numbers and not just vibes. And again, it will be good to know what's your total market. Right? What's your forecast? Because if you raise at a high valuation, it actually comes back and bite you.

Nathan Latka

20:04So are you able to confirm you guys were greater than a 15x multiple?

Jonathan Adly

20:08Yeah. Yeah. Easily. Yeah.

Nathan Latka

20:10And to your point, it's a good one, there is such thing as too high evaluation. So what multiple is too much? Like how do you figure out what the cap is?

Jonathan Adly

20:18When I talked to the Elephant folks, we sat down and we said, okay, let's make a vision. How could we hit 100,000,000 ARR? And we have a good relationship. Sometimes people don't have good relationships with their investors, I do. And we sat down and did the napkin math, and we'll be like, okay, so here it is. Does that valuation make sense if we execute on this forecast? And is the plan to execute on it realistic? And

20:43if it is, then great, the valuation makes sense. But if you're sitting there saying, you know, I think when SpaceX went public, they said like the total addressable market is like all the money in the world. Okay, you know, that's so that's what I'm saying is like, but I can say I have 450 pharmacies today. I know there are 16,000 pharmacies out there, right? I can go from 450 to a thousand, to 3,000. Great.

21:11So easily I can backtrack into the valuation from that.

Nathan Latka

21:14Yeah. 16,000 pharmacies with 2.5 bots each at 45,000 per bot per year is 1,800,000,000 of ARR. That would be a wonderful place for you to hit.

500,000 Prescriptions Per Day

Nathan Latka

21:28Okay, I imagine one of the usage metrics you track is how many millions of prescriptions are your bots filling every month or year. What's that number?

Jonathan Adly

21:35We are doing... Last we checked about three months ago, about half a million scripts per day.

Nathan Latka

21:39Your bots are processing half a million of prescriptions per day?

Jonathan Adly

21:42Yeah. And that's prescription. Bots are doing other stuff, but prescriptions are easy to kind of wrap your head around.

Nathan Latka

21:49That's wild. Well, look, I think a lot of people are going to want to follow your story. If they want to follow you online, where can they find you?

Jonathan Adly

21:55The best way is LinkedIn, you know, Jonathan Adly. Are very, this is LinkedIn is our outbound channel. The team does great work there.

Nathan Latka

22:05Guys, there you have it. He comes from years in the pharmaceutical space as a coder, as PhD, as a doctor, you know, got all the good stuff, launched TJM Labs back in 2024. Revenue as of recording is about $25,000,000 of ARR. Again, that's up 10,000,000 just in six months, an incredible growth rate. Ultimately, the goal is to scale to additional pharmacies, help them get more efficient in this age of AI. Jonathan, thank you for taking us

22:30to the top.

Jonathan Adly

22:30Thanks, Nathan.

Nathan Latka

22:32If you enjoyed that interview, you should click here to subscribe to the channel so you don't miss new interviews every Tuesday. If you've got some free time and wanna listen to the next interview right now, click right here. Lastly, subscribe wherever you get your podcasts. We're on Spotify, Apple Podcasts, and here on YouTube.