for Global Healthcare Conference. My name is Sean Dodge. I'm the healthcare IT analyst here at RBC, and I'm pleased to be joined by our next presenting company, which is Augmedix. And here on behalf of the company is their Chief Executive Officer, Manny Krakaris, and Chief Financial Officer, Paul Ginocchio. I got that right. You got mine right. You can get his right. So special, special thanks for you all making the trip out to join us here today. Just to make sure everybody's familiar with what Augmedix does, I want to have Manny just spend a few minutes at the beginning just to kind of give us a little bit of an overview. Sure. So, anytime you go see a doctor, the doctor has to perform the examination, we call it the encounter, and that, when it's completed, that doctor then has to document that encounter and deposit that documentation into the your electronic health record. That process consumes up to three hours of the doctor's time every day. As a consequence, it's imposing a great burden on them and leading to fatigue and burnout, which is a serious problem in healthcare. It's exacerbating some mega trends, which is that you've got a shrinking population of clinicians in the U.S. who are trying to accommodate a growing and aging patient population. So how do you address that gap, that disparity between your capacity to provide care and the demand for care? Well, technology is one way to do it, and we, Augmedix, came up with the idea of repurposing the conversation that occurs between the clinician and the patient, using that as the input for generating the medical note. We did this 11 years ago. It was revolutionary. People thought we were nuts, but now, fast-forward 11 years, the entire industry has embraced what is now called ambient documentation. It is being fueled by the advent of AI, which is facilitating the automation of the medical note when it goes through the process of creating a medical note, and I'll go through that in a moment. So the industry has rapidly adopted ambient, and we, as the pioneers, have had a head start in that space. We have enterprise accounts with 5 of the 10 largest U.S. healthcare enterprises. We have documented more than 10 million encounters, more than anybody, using ambient documentation, and that really is helpful when you're training your models, your large language models, to extract, relevant information from the transcript, which is the conversion of the audio recording of the conversation between doctor and patient, into a structured medical note. That gives us a significant advantage over anybody else who's in the space. So that's the background. I was inspired to join the company six years ago by my doctor. My doctor was, and still is, a customer of Augmedix, and even though I didn't pay any attention to her description of what was going on when she walked in the office with, at that time, Google Glass, I did remember her reaction to it. She said, "This has changed my life." And I thought, "Wow, that's pretty amazing to share with a patient, but okay, cool. Mm. Then maybe a year and a half later, I was invited by the board to join the company, and I was truly inspired by that memory, and heard similar anecdotes from many other customers. So the company had already established, by the time I joined, a very strong product market fit, and that's the toughest thing for any young business to get right at the outset. They had that. The next part of the journey was to scale. How do you scale rationally, and become a low-cost provider? And I knew that we had to use automation software to get there, and we're there now. So with that, I will dive into some details. Well- Next one, next one. So our product portfolio. Our legacy product is Live. That's how we started out 11 years ago, and it's basically a virtual scribing business where we assign a person, a medical documentation specialist, to observe the interaction between the doctor and the patient remotely. Most of our medical documentation specialists are overseas, primarily in India and Bangladesh. And they observe the interaction, and they generate the medical note with the assistance of a lot of tools that we've built for them to make that job relatively easy. We have substantial operations in those countries to recruit and train the requisite staff that we need to do that. That product accounted for roughly 90% of our revenue up until the end of last year. Now, our view was that that product, with the advent of AI would be sunset at some point, or at least wouldn't grow at the historical clip of 45%-50% a year. And so we developed other products to complement that product. As you see there, the intermediate product is called GoAssist, which we just launched a few weeks ago. And that is a pure AI product where the clinician has the option to invoke the assistance of one of our medical documentation specialists on demand. So if they happen to have an encounter that is particularly challenging, or if they're pressed for time, for example, I don't have time to document this, push the button. The note that's generated automatically by the back end is sent to the MDS. The MDS does the, what we call the last mile, fills in any gaps that the LLM technology may have missed, QAs it, and puts it into the EHR, the electronic health record, for the doctor to have a quick review and sign off on it. And then at the far, left there is Go, Augmedix Go, which is a pure self-serve product. There is no human intervention on our part. It is the lowest cost price product that we have. And there, the doctor just opens up the application, the encounter is recorded, it goes through our back end, which is an ASR, Automatic Speech recognition module or modules, and then through a series of large language models to generate the draft medical note for the doctor to QA review and sign off on. Pricing, roughly $2,400 a month for the Live solution. GoAssist is roughly at $600 per month, and the Go product is at roughly $300 per month. So we have a product for every price segment in the market to satisfy every particular need of clinicians and enterprises, because not every encounter is a routine follow-up encounter that is more appropriately documented through a pure AI product like Go. So we feel that we have established very good positioning in the market, where we can address all these, these varying needs of these enterprises. And these are built based on feedback we've gotten from all of our major customers, and we have relationships with 5 of the top 10 and more than 20 big enterprises. So competitive advantages. I kind of talked a little bit about these, but we've done 10 million-plus encounters, and we've developed through that deep domain expertise, knowledge across a variety of specialties. That allows us to deliver notes that are very, very accurate, relative to anybody else. We have the broadest portfolio in the marketplace, as I just described. Nobody else does that combination of pure AI supplemented with human intervention if the customer wants it. It's either one or the other. Nobody has both. Importantly, when we started this company, we had a bidirectional communication channel, which is important because that was the only way that we could deliver the Live service. Well, we're leveraging that now by delivering third-party data and information to the point of care at the appropriate time during that encounter. For example, in the emergency room, if a doctor encounters a certain condition of a patient, we can deliver an appropriate prompt to that doctor as to what to look for next, what to order, what to examine, based on the cues we're getting, the ambient cues we're getting from the interaction in that emergency room between the doctor and the patient or whoever's representing the patient if the patient can't speak. That is revolutionary. That did not exist before this, before our ability to do that, and that's of high value to our customers, in particular HCA. As I mentioned before, we're deployed at multiple major enterprises, and I can't read that last one. Oh, yeah. We've got infrastructure that we've already built. We're not a product company. It's not like we just built a product and people are going to flock to the product. You have to have the infrastructure to support that product. We've been around for 11 years. We've built that infrastructure. We are HITRUST certified. We can support every product that is deployed at our customers very well. So, we're full guns ablaze. That's it. All right. So we covered the competitive landscape. So it's fairly dense landscape, lots of competition out there, but as you pointed out, you're aligned with some of the biggest and best health systems out there. Got a lot of different competitive advantages, attributions. If we maybe then turn to Q1 and the results that you reported yesterday. You did see some customers shifting from Live into GoAssist. So you said you have recently launched GoAssist and you're already seeing some trading. Maybe just talk about that dynamic and Sure. Well, some healthcare enterprises are under severe financial pressure to improve their operating efficiency. And we took the initiative of going to those customers and telling them about GoAssist, and it's not quite as good as Live because you don't get the ancillary services, like referrals and pending orders, and coding, and so forth, and nudges to the clinician that you do with Live. But if they can do without those ancillary services, we could deliver a pure AI product to them with some support from our MDSs at a much lower price point. And therefore, you could deploy it much more broadly within the enterprise, and that resonated with these customers. We immediately got requested: "Hey, okay, we wanna transfer X number of doctors from Live to GoAssist, like, as soon as you can." And therefore, frees up more revenue dollars, the ones that they were already spending, they could deploy towards more doctors, so they can provide relief to a greater percentage of their doctor population than would otherwise be the case if they are so budget-constrained. We think that maximization of their customers' ROI is gonna be good for us long term. Okay. So Manny, you said it sounded like there was just a few kinda isolated instances of customers that were under pressure. I guess, what gives you confidence at this point that you kinda have your arms around this kind of product mix dynamic? Well, we're getting a lot of inbound interest. We haven't publicized it- Okay ... I mean, to the customers widely. We've talked to our existing customers about this product, but we haven't launched a marketing campaign to promote GoAssist. People know about Go, but they don't, and they know about Live, they don't know about GoAssist, that intermediate product. We think it's gonna resonate because as providers trial the pure AI product, I think there's gonna be a bit of a misalignment between expectations and reality. Because the pure AI product, I don't care which vendor is offering it, doesn't deliver against those expectations that have been, I think, perhaps a little bit too aggressively set with healthcare providers. And so they're gonna look for something else, and if that something else isn't that much more expensive and still delivers a compelling ROI, and I'll talk about that in a moment, they're gonna be interested. Mm-hmm. So we talked about your alignment with, with some of the biggest and best health systems out there. HCA is, is one that you have a, an agreement with. But maybe just talk about what you're, what you're doing with them and, and how your work with HCA is progressing. Sure. So HCA came to us a couple of years ago and asked us if we would be able to build a fully automated product for the emergency department, which is a very, very challenging environment. You have a nonlinear workflow of the clinician, everything is unscheduled, so you've gotta have a technology that is able to follow a doctor while that doctor sees multiple patients throughout that shift, and they're not sequential. They're. The doctor goes back and forth based on results that they might order in terms of labs or blood tests and so forth, radiology. And so, it's a very, very difficult product to build, but we had had experience with our Live solution in another emergency department of another hospital, another big enterprise. We had learned how to master that non-sequential or nonlinear workflow challenge, and we felt confident that we could build this fully automated product for that environment. And we showed them, you know, our technology stack. They spent several days at our offices, their entire team of data scientists, CMIO, and you name it. There was a big, big crew for three days, and we went through everything, and they evaluated us against Microsoft, Nuance, and a couple of others, and they chose us. We've been working at developing that product. It's being trialed today by several doctors at four hospitals, our HCA, and the feedback we're getting is very, very positive. Importantly for HCA is: What is the impact on our economics? What kind of an ROI can it deliver? We just got the latest compilation of that data this week, and it is extremely strong. Mm-hmm. We're very, very encouraged by the data. And so they are fully committed to deploying this broadly across their enterprise, the ED product. But beyond that, there are multiple growth vectors within HCA, different departments on the clinical side and even on the hospital side, that they are... We've already started, you know, working on deployment plans, in those other areas. And it's all driven by: What impact do we have on their wRVUs? Mm-hmm. It is very, very significant. It's interesting how you save doctors an hour of time, how that can be repurposed in terms of wRVUs, but also capturing everything that is reimbursable in the note. That capture rate is very, very powerful. So to me, the hard ROIs, they're being driven by physician productivity. You're improving that, you're capturing more billing opportunities. But I would imagine it's not capturing the qualitative improvements to where you're just you're helping with physician burnout, which is something that- Exactly right ... most systems have really struggled with the source. Exactly right. It's a pervasive problem throughout the industry... And that is cited as one of the biggest pain points, and one of the biggest reasons why this technology is being adopted so quickly now. Manny, you mentioned, so you're piloting it now or building it in the emergency department with HCA. There's opportunities to expand across all of HCA's hospitals, so outside of just the four that you're in now. Maybe just talk a little bit about the other opportunities within HCA. I think you mentioned other care verticals, other- Mm-hmm. Other care sites. Sure. So the clinical side is much bigger than the emergency department side, for HCA. HCA has a total of about 45,000 doctors, and about 4,500 of those are in the emergency department. The rest are on the clinical side, or hospitalists, et cetera, in the hospital settings, inpatient settings. We have been deploying small cohorts, on the clinical side, and we also got a first wave of data on GoAssist, in those areas. And again, the economics are truly, truly compelling. So we're very encouraged by that, and it's led to more meetings, by these business units. There are, I think, eight or nine different business units outside of the hospital group, that we've had meetings with that are interested in deploying this product. Okay. And then if we turn back to guidance, you did adjust revenue outlook a little bit for the year related to the dynamic we talked about with the customers trading from Live to GoAssist, but didn't change the OpEx guidance. Maybe just your thoughts on that. Sorry, didn't change the what? OpEx. Oh, OpEx. Yeah, I see. Oh, sure, yeah. What you expect to spend? Yeah. So, as I mentioned earlier, we had anticipated that there would be a shift in demand away from taking away the growth of Live and into the growth of these lower price products. We didn't expect it to happen so quickly. Mm-hmm. And so when you have a transition like that, where you've got users going from one product to a different product that's priced substantially lower, until you get that volume to compensate for the price adjustment, you're gonna see that reduction. And so we've seen that. On the OpEx side, we've as I think I stated in my prepared remarks on Monday, during our earnings call, we've committed to about 60% of the hiring plan for 2024. We will see how market, you know, sentiment, not investor sentiment, sorry, but, our customers react to the offerings that we've just recently released. And if we see indications that things aren't gonna ramp up as fast as we thought, then we'll make some adjustments to that, of course. I mean, that's what we get paid to do. We manage the business. And if we think about the first quarter, cash burn was about $8 million, how should we think about the pace or the cadence of that over the remainder of the year? Yeah. So, first quarter is typically our biggest cash burn quarter of the year due to annual compensation payments, annual software payments. So that would be a high watermark. I think maybe the best way to think about our change in cash this year is to start with our, obviously, our revenue guidance of $52-$55. You know, look at our second quarter gross margin guidance, and, you know, our gross profit would be just under half of that revenue guidance. We talked about OpEx being, you know, 12, mid-50s, and you could. So, that takes you to a net loss. I think people need to remember that adding back to that net loss would be about $4 million of stock-based comp. We'd also get a couple million of positive net working capital, and then we also still have $5 million left on the facility. So I think if you did all that math, using our public guidance, you would get to a change in cash for the year after taking down that last $5 million of the facility, of $20 million or less. Okay. Then you do have, is it $20 million of debt that have payments that will potentially start in July? How has that been factored in? Yep. Great. Yeah, good, good question. So, we're currently in active negotiations with SVB to make sure that our IO period, which in the facility can go out to is the middle of 2025- Mm-hmm That we have that whole IO period to the middle of next year. Yeah, we negotiated that when we first did the deal, that we could extend it to July of 2025- Okay the beginning of repayment. And, we're in the process of making sure that we get there. And then, if we go back to the dynamic that we talked about earlier, the trading from Live to GoAssist, Manny, if you could just give us an idea of what the margins are. So you gave us the price points- Mm-hmm ... what the margins are of those different products. And as you launch Go and get traction with that, and you launch GoAssist and get traction with that, what does that do to your gross margin profile? Sure. Well- We've always stated publicly at scale that Live would have gross margins of between 50% and 55%. Okay. We're pretty close to that now. Go would have margins in excess of 60%, and then GoAssist would be somewhere in between, 'cause there is a human involved in that, so the cost would be a little bit higher. So you can kind of go from there. Obviously, we're gonna exit the year with more exposure to Go and GoAssist. It'll be much more meaningful to our overall revenue base and bookings base next year. And that will, well, on the revenue side, drive that gross margin higher, depending on the mix. Yeah, and I appreciate it's still early in launching all of these. You said Live was about 90% of the mix- Mm-hmm as of the end of last year. Correct. Where do you see, if we roll this forward a couple of years, what would you expect the portfolio to look like, mix-wise, across the different solutions? It depends what year you're looking at, but if you're looking at 2026, we might see Live down to 50% or less of our revenue. All right. Well, we can, we can leave it there. Great. I appreciate both of you, joining us again. Thank you. Thanks a lot, Sean. Thank you.