The Doctor Who Left Medicine to Rebuild Its Rulebook with Christian Chartier CEO& Co-Founder @ Compiled Health

[00:00:00]

Nectar: Christian, pleasure to have you on the podcast. So much to chat about, man. I'm so excited to learn more about Compiled. Obviously, we've known each other, we've known each other a little bit before I've got a bit of an inside view as to what you guys are doing, so it's really curious to dive in more.

Nectar: Maybe I'll start with just a very simple question is what is Compiled Health?

Christian: Yeah. So I guess it requires the background a bit about me as well. So physician by training, and, for obvious reasons our healthcare system is quite strapped. And as early as residency we were confronted with that face to face.

Christian: And tangibly, in, in family practice and primary care, what that meant for us was, very short visits. You're expected to, from memory, recall infinite rules of care, the things that justify why you offer tests to patients, who you offer them to, at what cadence, and based on what kind of grounding. W- we found ourselves trying to quickly, [00:01:00] in a very unstructured way interview patients, fact find from them as quickly as possible, and then with what's left over in a, a 12 or so minute visit counsel a patient and hopefully lead them to providing some sort of informed consent about what the right path forward for them, whether that's ordering a test considering a new medication, considering stopping a medication.

Christian: And what we began to wonder is, what a more structured version of that first fact-finding piece might look like in order to free up some of the remaining time, albeit still quite short and give that back to the patient, meaning spending more time with them, laying out the options, making sure they truly understand the pros and cons of each.

Christian: Classic example is in the space of screening. When you're considering screening for a certain cancer or chronic disease, there are risks and benefits and patients may not be aware of them. And we found that part of the conversation, which of course is critical and grounded in the evidence itself, was being totally crowded out.

Christian: So when we began [00:02:00] examining the evidence that we were supposed to memorize as doctors and other professionals we wondered what it might look like not to be consulting them in PDF form, but for them to be transformed into something more usable that could support the first part of the, the interview, so the fact finding, and then leave time for the rest.

Christian: So what Compiled is that very process. So it's the pipeline designed to intake the medical wisdom, so consensus documents, guidelines drafted by thought leaders in the various spaces in which they operate and publish and meant for us to consume and apply. And Compiled transforms them into machine-readable artifacts, which you can think of as kind of decision trees that ease the first part of the interview to free up time for the second and critical part of being able to counsel patients, really advise them and elicit informed consent so that they ultimately do what's best for them and that they're aware of the trade-offs

Nectar: So if I break this down a bit the way it works today is I go see a doctor, I go see you and you're like, "I'm not feeling well," and y- you have to memorize a bunch of [00:03:00] different if this, then that kind of questions, and that leads you down to make a, a diagnostic hey Nector, I think you have X as a as a condition.

Nectar: So what you've made this has made this all legible to the machine, in essence, right? So the... All these questions are posed in the meeting to your system, and then that helps basically, the, the physician or the nurse quickly get to the conclusion, right? So just bypass all this stuff that's prob- I imagine fraught with a high degree of error and get to something much more accurate.

Nectar: I- is that a simplistic way of explaining your product?

Christian: Yeah. The kind of analogy I began using, which is the, the childhood game of 20 Questions, and so you'll be playing with somebody and you'll say, hey, Nector, guess any animal." And then once you've guessed the animal, by no more than 20 questions can I, with 100% certainty, guess the animal you're thinking of.

Christian: And so it'll be like, does it swim? Does it have wings? Does it have a beak? And as you tease apart the various options and eliminate the vast majority of them, you do [00:04:00] arrive at a knowably correct right answer that they've guessed. And so- Our first use case being screening. If you think of colorectal cancer screening, for example, there is the choice to test or not to test, the means by which to test, and the timeline according to which to test, and these are all folks that similar to twenty...

Christian: These are all features that similar to twenty questions people have about them as an innate property. And what we need to do is structure the interview and ask the right question at the right time to eliminate as many of the kind of incorrect paths forward as possible. And so as a first step, what we did, exactly that for cancers and chronic diseases, is take in all the evidence, all the guidelines, merge them together, and turn them into one master decision tree that allows us, in as few questions as possible, arrive at the correct next step across all cancers and chronic diseases grounded in the features elicited from the patient.

Nectar: So there's no probabilistic AI in [00:05:00] the front end, let's say, of the process. So where does AI play a role then, right? Is it really to gather and ingest the tens of thousands of documents, research, staying updated? Like, how does... how do you build your product and ma- then make sure that it's deterministic in nature and there, there's no errors?

Christian: As some of these new tools gain traction and begin being used in everyday clinics doctors are still held to the standard that their license prescribes. So if you make a clinical decision yourself or you outsource it to AI, what you're held to and maybe asked to call, recall or talk anyone through months removed, is the decision you made, how you documented making it, and what were the, what was the kind of underlying reasoning that led to it.

Christian: And so using AI in a chatbot form at inference time, so with the patient in front of you, doesn't so much allow you to do each of those things the way you would if you were making a decision or reasoning over it yourself. Where the AI does come into play [00:06:00] is far removed from the patient at the ingestion time of those guidelines.

Christian: So an AI can rapidly take a full library of medical guidelines, reason over it, architect a tree that is high fidelity, meaning it captures all of the kind of details and logic encoded in them in plain English, and turning them into something that at inference time, so when you're in front of a patient, you can know will reliably lead you to the right answer one hundred percent of the time, whether you did it on Monday, whether you do it on a Friday, or whether you need to defend it some months after.

Christian: So the AI is actually building the artifacts. The AI can also traverse the artifacts and administer them, but by no means is there AI to what we do when the patient's involved.

Nectar: Yeah. Like I wanna unpack the, maybe the product and the vision a bit 'cause I find this fascinating.

Nectar: There's so many parts of our healthcare system that's broken, particularly in Canada, but I imagine, and we're not the only country that has this. I wanna maybe just dive more into your personal history, Christian, if that's okay. It's like how many years did you study to become a doctor? [00:07:00]

Christian: Yeah. We're fortunate in Quebec and at McGill in particular, where there is a fast track into and through medicine.

Christian: And so what that means is effectively after grade 13, you can enter a five-year program which license you as a physician much quicker than traditional, undergrad followed by medical school and so on. Fortunate to be admitted to one of those cohorts. I began when I was 19, but very quickly realized two things.

Christian: I think the first was I had no business being in front of patients at 19 or 20 years old. A- and secondly, definitely took an interest in technology, but felt that the kind of intersection of health and technology at the time, specifically on the health side, was very descriptive. There wasn't so much building happening as there was describing of things going on, talking about new technologies rather than building them.

Christian: And about halfway through I did make the decision to step away. I took some time away and went to study maths and computer science, and the thinking was that upon return to medical school, would be able to work on some of the same [00:08:00] things, but with a new set of tools to be able to contribute more meaningfully.

Christian: So in this case, what that meant was be able to build as part of a research project rather than describing existing technologies. Over time, some of those kind of small research projects ended up getting spun out in particular in the image generation space ultimately selling image generation tools to plastic surgeons, so both in the aesthetic and the reconstructive space, this idea that a patient could submit a preoperative picture, and our tool would spit out what they could plausibly expect to look like after.

Christian: And this was, in some sense before AI was cool, and the impact that we felt we could have was ultimately eliciting, again, a more informed consent from a patient as to what it was they were signing up to do. So that was the, the, the first foray into research translating into technology, and ended up graduating about a year ago from residency and haven't looked back since.

Nectar: Yeah, I was gonna say, man I can't imagine the, the, the conversations with your friends and family. It's like, "Hey I studied to be a doctor. Now I'm going to [00:09:00] start a startup instead." But you didn't directly start a startup, but you've had a few paths before, right? If if I'm not mistaken.

Nectar: But we could talk about OpenAI and your, a little bit of your time there, but I'm more curious about, that, that decision, right? That you said, "Okay, I'm gonna screw this all the time," and obviously you're using this knowledge, but you're gonna focus your efforts more to build a technology that's gonna hopefully benefit millions of people.

Nectar: Walk us through that a little bit, man. I find this such a brave thing to do, not easy- Considering all the time you spent studying. Was it a hard decision or is it just naturally you're like, it's more your personality type of "No, I made the call and that's it," and rip the Band-Aid?

Christian: Yeah. D- totally telling question. I, I don't come from, necessarily any, any means, just a kinda nor- normal middle class background. Th- there was this anecdote that stood out during residency. Antoine, my co-founder, and I realized that over a career, a physician can expect to see some tens of thousands of patient visits total, and that's it.

Christian: And as, as technology is kinda taking off and as [00:10:00] this thing that we call AI is becoming clinician-facing, we figured there might be a path to having that same impact, but much quicker. And I can't mention enough how, I won't say de-risk, but de-risk in a sense, the whole step was to ultimately leave clinical medicine, 'cause I was able to do it with someone I respected, someone I knew training on the other side, and who had agreed to take this leap with me.

Christian: With respect to the numbers, we can kinda go back to those. In six or so months of launching in a few sites, our software has actually already emitted some 20,000 clinical recommendations. So maybe it's a question of ego or maybe it's just the, the ways by which Antoine and I define impact, but we kinda have that number on the screen, and every day we look to the, the number that we would've seen otherwise, right?

Christian: The number that we'd be seeing in kinda hand-to-hand combat with patients. And at some point, that threshold will be crossed and exceeded. We hope it's soon, but we know it will come at some point. And ultimately, it does come down to the impact and it feels that our time and effort [00:11:00] and e- emotion and blood, sweat, and tears and all these things are best invested doing exactly this at exactly this time with the patients who benefit from it in mind.

Nectar: That's super inspiring. Super inspiring. There's hope for the rest of us. You-- I wanna go back to the product a little bit. You mentioned this 20,000 KPI. How do you-- how does the customer value it on their end, right? So is it how do they track? Is it time saved? Is it reduced, what's the...

Nectar: I'm looking for, blanking on the word, but, when people come back to the hospital, readmissions, right? So what-- I imagine there's more to it than just the, this, this specific KPI that you mentioned, right?

Christian: Yeah. A- absolutely. We are operating cross border, and Antoine and I are both trained in practice in both settings, Canada and the US.

Christian: And the incentives are aligned differently for obvious reasons. Now where we began, which is in the preventive care space it made sense as a first step for a few reasons. First preventive care guidelines are actually meant to be applied directly, right? They're ap- meant to be applied at population scale.

Christian: They aren't necessarily [00:12:00] handed, packaged up and handed to physicians, but they are spread out across the full continuum of healthcare professionals. And what that meant is that there were actually already clinics that were practicing preventive care in kind of the style I was describing it before.

Christian: This was all the, the repurposed COVID testing and vaccination apparatus was being converted into prevention centers. And what we found there was nurses doing largely what we hoped that our software could one day do. They were doing the same things and encountering the same failures that we were in clinic, and their goal was actually to just take the guideline and apply it directly.

Christian: So it was the, the perfect alignment of a well-defined set of guidelines, existing workflows that we could work with and seek to optimize and we're familiar with. And ultimately, that, that is the conduit through which we've been able to emit so many recommendations. Now, in the US, there's a bit of the same.

Christian: The- there, there are some false beliefs about the US healthcare system, but specifically as it applies to prevention and screening, there's a set of US preventive task [00:13:00] force, what they call Grade A and B recommendations, which are similar to the screening protocols that we follow here in Canada.

Christian: The findings of those consensus documents, meaning the recommendations that they're designed to support, by law need to be provided to all Americans for free. So in a sense, this is the niche of public access healthcare out in the US. And so the natural analog to what we're doing here is to rinse and repeat, ingest those same guidelines, and distill them to artifacts for prevention down in the US The incentives are aligned differently in in other parts of the, the US healthcare system, namely as it pertains to insurance as it pertains to kind of revenue uplift and billing.

Christian: And these are very crowded spaces that we're not gonna pretend to be able to upend the leaders in. But there certainly is this notion of kind of conditions needing to be met in order for an action or a billing or a reimbursement to take place, and these are massive pain points when they fail.

Christian: Hospitals have reimbursement rates for a reason, and of course, their goal is to [00:14:00] maximize them, and insurance companies do fight back. Our goal is that a patient who meets conditions for a treatment, a medication, a procedure, not be denied it or delayed that care because of some administrative lag built into a system.

Christian: Rather, if they deserve the care, they should be offered it. And our goal in inserting ourself in the US healthcare system is to ease just that and smooth that, that process.

Nectar: Yeah, it's interesting 'cause that-- it makes me think that your business model has to adapt per region, public or private, right?

Nectar: So already multiple fronts. Maybe at a broad level, what have you learned now that you're on the other side? You're obviously not a not a practicing clinician anymore, but now sailing into that space. Are there any key takeaways for people that are in health tech?

Christian: Yeah, so one piece of guidance we got early on was don't waste your time trying to work with government, and don't waste your time trying to sell into government. And implicitly what that meant is that the work that you guys are doing cannot [00:15:00] benefit the population who supported your training, the systems themselves that trained you and who you arrived at some of these conclusions and brand new ideas based on.

Christian: And that ended up being bad advice at the time. And being able to work within the, the means of the public healthcare system has paid off for us. Staying true to this mission of the patient who qualifies for the care, who needs the care, who meets criteria for the care should have the care, should not necessarily depend on their ability to pay, the fact that they happen to be in a private hospital over a public one.

Christian: And ultimately, we've been able to adjudicate and process these rules of care and these guidelines in a much more like business neutral way by handing the artifacts off to a system that trained us, and obviously the, the impact that th- that can have the kind of equity and equality to the downstream impacts and so on is quite clear.

Christian: That said, versions of this same tool can be used across the full kind of health tech stack. And I would encourage, anybody who's getting into [00:16:00] this at the same time as we are, t- to think meaningfully about the kinds of impact that they'd like to have and what limitations they may be unknowingly acquiescing to by working with a narrow part of healthcare or just defaulting to the easy part or to the patients who are able to pay.

Nectar: Yeah. Yeah, you're definitely doing it in hard mode . That's for sure. How do how do you then, the objection of hypergrowth, right? Growing quickly. How do you plan to counter that? Is it by just simply having a multiple footprint, like you said, like going through har- what's expected and water flows to the easy places like you said, "Oh, j- just don't do that, do the hard stuff."

Nectar: Is it just a balancing of different go-to-market motions?

Christian: Yeah. So emerging from fundraise mode I'll take the hat off. We've had to like tirelessly explain how our approach and our architecture and our premise is different from, the whole cohort of really amazing clinician and patient-facing healthcare technology companies that are out there.

Christian: And when you think of growth, [00:17:00] once we can take that hat off, we actually view these folks as potential customers, potential conduits to the intelligence that we can package up. It's important to note that our first deployment with, an end-to-end web application serving multiple healthcare systems was not actually our core competency.

Christian: Our core competency was packaging up the intelligence in a form that was more useful to the clinician or could have a more direct impact on the patient. And that doesn't necessarily require that it be pushed through our channels, right? When you think of an open evidence experience with OpenAI and the kind of products that, that use OpenAI models, like we actually key in quite nicely with them.

Christian: We don't view them necessarily as competitors. And if we can support a, chat-based AI for health s-software suite with more grounded and more rigorous evaluations of the rules of care then we'll do that, and we expect to be to see infinitely more impact that might come with it.

Nectar: Yeah. I imagine on the superficial level, everyone knows that, health is an issue. People get sick, right? And like it's all going the right direction, and so I don't think you need to [00:18:00] explain the TAM slide, right? But I'd be, still be curious to like how do you frame it, like in the sense of it's such a common thing that people get sick.

Nectar: Is there a just a question of it's aging population or is there another like the prism through which you look at your market?

Christian: Yeah, so there is a question of this pie can get much bigger. Henry Ford years ago said how he was considering manually making some of his cars versus, thinking through what the proper factory floor might look like and how to get those cars out to, today hundreds of millions, maybe even billions of people.

Christian: And this idea that doing it bespoke by hand, maybe making one or two cars a year didn't function at scale. Not only was it bad for business, but it was worse for people. And what we seek to do with this is rather than y- a doctor's kind of training and their thought process and the full skill set that they've developed over their training and their career be restrained to the patient that they can have, directly in front of them and how quickly they can shuffle them in and out of the [00:19:00] room, for us is a false premise.

Christian: And by, A, widening access to information, which, other tools like ours are doing, but also guaranteeing that as that information propagates through new systems and new technologies, that the quality control bar is set just as high. We think there are many folks who are working on solving the first part, making the information accessible.

Christian: Any, any doctor will tell you about a patient who's gone through Dr. Google when they've, walk into your office. But knowing how to walk them to the knowable right answer when there is one through a consumer-facing product, for example, is a totally separate challenge and one that we believe is a necessary condition to actually widening access and broadening access to care.

Nectar: Yeah. Does the role of the doctor change, do you think? Does the, 'Cause the way you think about it today, it's like a doctor spends all this time to ingest the knowledge, stay up to date, and then, provide that care to the person. But the algorithms are gonna be doing this today, right?

Nectar: How does the, the role of that person that, you know or obviously the nurse as well, like, how do they, their roles change con- considering [00:20:00] that, we're getting the AIs in the rooms whether we want to or not, right?

Christian: Yeah. Part of part of wanting to do this is a fear that I personally experienced, which was almost this idea that the doctor's getting left behind.

Christian: Our medical systems are organized around a license to practice, and what we're seeing today is more and more tools exposed to those license holders. You're expected to use chat, you're expected to use clinical decision support, while we simultaneously see the deference in their ability to audit the ways in which those tools arrive at the decisions they emit completely atrophied.

Christian: And of course, today, most doctors haven't been sued, and you'd like to hope that most outcomes are quite good in the healthcare system. But as doctors ultimately do, in some sense, get left behind they may one day be called on to explain the reasoning or to, walk, God forbid, a court, but potentially a colleague or a patient themself through a decision that they made [00:21:00] some months ago.

Christian: And previously what we did was ask that decisions be documented, that reasoning traces be written in the chart. But when you look to chat and the kind of tech space that's organized around it, you don't so much have that auditable and version-controlled reasoning trace. A great example from last week was the Astra release and the model card that came with it, and this fear that- The reasoning trace that is sent back to the user may itself be an artifact managed by the model, where it may not translate one-to-one to the actual underpinnings of the response you receive, and that it might just be some separate facade.

Christian: And, as models reason through different frameworks and become more and more capable, if that reasoning trace, even if it were perfectly reliable, gets lost then we've totally lost the plot. So part of wanting to do this is reframe the conversation about the license to practice medicine, what bars are set, and what standards are expected of [00:22:00] those license holders, and how can we organize a system both around this new technology but without forgetting what it means to omit a medical recommendation or do a medical act.

Christian: And in some sense, a bit rambling here, but in some sense a video producer, of course, benefits from video models being able to generate kind of quick clips of video and do plausibly looking human-generated. But they still do have their kind of, film and editing studio that takes place outside of the, the kind of live generation and the the, the rapid fire generation of films where they can actually sit with their media, sit with their content, reason over it, adjudicate some issues that they might find and then ultimately make the decision to publish something or to, to put something out on the big screen.

Christian: We... Certainly in medicine, that kind of suite is what we're looking to build. We want folks to know that the tools that they're using when patients are involved adhere to the standard that's set for them when they choose to take their license and when they choose to practice medicine. [00:23:00]

Nectar: Yeah. That's a fascinating, conversation, Christian, about, like, how you see the future of healthcare, and you're really at the forefront of it.

Nectar: And and obviously maybe to talk about Fundraise a bit, congrats on closing your round that was just announced, so well done. Maybe on, on the company building side, like, how do you see, let's say, the next 12, 18 months now that you have some some more firepower b-behind the company? Like, how do you plan to basically take the company to the next level?

Christian: Yeah, a-and thanks for that. It was a, a really surreal experience. I'd never really done anything like that and neither had my my, my colleagues here. So it's a really exciting process, a humbling process. I think what the fundraise here unlocks is it permits us to go from where we began, which was a kind of product application layer where we wanted to build some usable set of features that clinicians were willing to use in real time.

Christian: And that was the conduit to this new form of intelligence and these kind of more usable artifacts representing guidelines making their way into patients' hands or providers' hands. Now- As we started [00:24:00] scrutinizing what it actually means to formalize medical knowledge, to formalize guidelines and standards, we realized that there was a much deeper study of language, of ambiguity, and of specificity that was required in order to do this right.

Christian: Now, what I mean by that is the way guidelines are written down, it's actually never for a machine, a computer, or an agentic audience. It was meant for doctors to read them with the proper context, the background, the training that the community expects in order to apply them in the way that they were written and intended to be applied.

Christian: Now, as you go and, try to make the-- all, all those machine readable, you're unknowingly taking positions on things that were purposely left ambiguous within the writing itself, and spending time, adjudicating the use of the word consider, the use of the word offer, the baked-in meaning, how it translates between, languages, in our case, French and English.

Christian: That is not the business of kind of an application [00:25:00] layer company or at least one that was poorly capitalized and was just kinda trying to bootstrap. So as we were convincing folks that we were different from, go-to-market wrappers for frontier intelligence, we found ourselves, like, wondering what actually doing that meant and how deep down the rabbit hole we might have to go on guidelines themselves, on translation of guidelines between languages and study that the work that came before us in order to do it right.

Christian: And so raising the, the round meant that we could separate concerns. PCare remains the product clinician-facing layer. Compile is a more nebulous kind of version now coming to life of that kind of proper study and that philosophy of wanting to do this at a much deeper level in order to do it right.

Nectar: And can-- Does PCare give you a wedge into future applications, future products in terms of everything that you're collecting? 'Cause you're obviously collecting all the, the, the medical side, but then you're also collecting, I imagine if it, if it-- there isn't this zero, zero data retention policy, but if you're collecting as to what doctors are actually seeing in [00:26:00] the front lines, right?

Nectar: Are there potential areas that you can create more value with your product?

Christian: Yeah. So a f- a few things there. As as the pace through which knowledge gets written in consensus documents and quickly propagated into direct patient care, as that quickens, the methods that guideline publishers and professional societies need to work with to better surveil the literature and to better surveil real patient outcomes also changes.

Christian: In some sense, data emanating from the direct application of our artifact should bubble up and be useful and those conversations are already happening. Now with respect to the first product being a wedge that, that might lead to others this whole space of kind of clinical AI began with the scribes that made every physician's life easier, or so it felt.

Christian: And, the, the metrics there were reduced physician burnout, reduced time spent documenting, reduced time spent after clinic. And so we naturally honed in on these AIs emitting language or [00:27:00] documenting faster or doing text-based tasks much faster. Now, this idea of recommendations or clinically correct next steps being the sum of the criteria being satisfied that they require applies elsewhere and it doesn't necessarily have to only be on the text basis.

Christian: Now an obvious next one is the idea of wait list optimization. When you refer a patient either to see a specialist or to see a surgeon or to some other part of the healthcare system, there is a scarcity to it. And reconciling who should be at the front of the line based on their needs, based on their situation is not obvious and in our view, not being done all that well.

Christian: So reconciling criteria to better order those wait lists and make sure that the folks who need the care faster do get it is definitely on the horizon. There's some crazy stat about, north of 50% of first time prescribed medications in the US actually being refused by private insurers. If that's a business principle that's been applied or the [00:28:00] kind of slowness of that process is by design, then that's one thing.

Christian: But i-in a sense, if a patient needs medication, and they meet criteria for it, our goal is to advocate for them using our tool and make doctors' documentation burden of doing that advocacy much more straightforward so that we can smooth that transition. So there's a kind of infinite set of pro- of products that emanate from like resolving criteria and meeting standards, and they don't necessarily all have to be on a text basis or in the form of a chatbot.

Nectar: Yeah, no, that's super exciting stuff. Man, I find it's kind of the future. Talk about the future, like we both, you know, like you studied obviously computer science. You're at the forefront of this AI wave. The thing that I wouldn't say scares me, but concerns me is the diffusion and adoption of, from traditional industries like healthcare.

Nectar: What makes you optimistic today? We can talk about the bad side and all that stuff, but I'm more curious on the positive side. Is there anything that makes you optimistic in terms of adoption of these new technologies in your job, right? You're selling to these people now, right?

Nectar: C-curious to see, what you're excited about.

Christian: Yeah, absolutely. During training I would, speak to folks who are not in medicine and just [00:29:00] ask them "Hey I know this cool piece of tech that you're working on in some other space. Like how might that apply to what we do?

Christian: Like, why aren't we seeing that same kind of like knowledge transfer happening between fields and ultimately improving the care, improving, patient access and so on?" And historically, what I'd heard is like, "Oh yeah too much red tape." The, the best folks don't really wanna work on that issue.

Christian: And since those early conversations and with the rise of some of the kind of more clinician-facing AI products like the Scribes, like we've actually seen a massive appetite within healthcare professionals, within patient populations and so on, for these types of tools because, like you said before, getting sick and needing healthcare is an inevitability.

Christian: And red tape should not be a reason, regulation should not be a reason for folks not to wanna grab these things by the horns and be early adopters of them, as we've seen some of the, the tools being today. So I'm optimistic about the marketing, the market really matching needs and wants here rather than it just being, chalked up to regulation or red tape.

Nectar: Yeah. No, it's super exciting, [00:30:00] Christian. I could probably bug you for a few more hours, but you're a founder. You have other stuff to do. If people wanna follow the journey, follow Compiled what's the best way?

Christian: Yeah. So we've spun up our website, compiledhealth.com. W- we've put out some pieces of writing which we feel best articulate our perspective and we're really excited about them.

Christian: Christian@compiledhealth.com is a, a best way to reach me personally. We're actively hiring obviously on, on both the research and more product traditional software sides both of which are equally exciting for us. So it's the best way to reach us, just reach out.

Nectar: Yeah. Thanks for your time.

Christian: Thanks a lot, Nestor