WEBVTT 1 00:00:02.080 --> 00:00:04.319 - There's a small circle of people who truly 2 00:00:04.319 --> 00:00:05.919 - understand what it takes to lead a health 3 00:00:05.919 --> 00:00:08.660 - system right now. The decisions are enormous. 4 00:00:08.960 --> 00:00:11.919 - The variables are constantly shifting, and the stakes 5 00:00:11.919 --> 00:00:14.099 - of getting a call wrong don't stay contained. 6 00:00:14.400 --> 00:00:16.824 - That's not a complaint. It's a structural reality. 7 00:00:17.204 --> 00:00:19.125 - And it's why the most effective c suite 8 00:00:19.125 --> 00:00:21.285 - leaders are deliberate about where they go to 9 00:00:21.285 --> 00:00:22.585 - pressure test their thinking. 10 00:00:22.964 --> 00:00:25.945 - Becker's fourteenth annual CEO and CFO roundtable 11 00:00:26.324 --> 00:00:28.564 - brings together more than 1,500 12 00:00:28.564 --> 00:00:29.864 - executive level attendees 13 00:00:30.239 --> 00:00:32.799 - and nearly 400 speakers for four days built 14 00:00:32.799 --> 00:00:34.739 - around growth, financial sustainability, 15 00:00:35.280 --> 00:00:38.079 - workforce strategy, and the leadership questions that don't 16 00:00:38.079 --> 00:00:39.140 - have easy answers. 17 00:00:39.679 --> 00:00:41.840 - Join us November second through the fifth in 18 00:00:41.840 --> 00:00:42.340 - Chicago. 19 00:00:42.799 --> 00:00:44.820 - For the agenda and event details, 20 00:00:45.195 --> 00:00:47.355 - visit beckershospitalreview.com 21 00:00:47.355 --> 00:00:49.115 - and click on the events tab in the 22 00:00:49.115 --> 00:00:49.935 - upper right. 23 00:00:50.554 --> 00:00:53.675 - This is Scott Becker with the Becker's Healthcare 24 00:00:53.675 --> 00:00:54.175 - Podcast. 25 00:00:54.635 --> 00:00:56.475 - I'm thrilled today to be joined by an 26 00:00:56.475 --> 00:00:57.454 - absolutely brilliant 27 00:00:57.835 --> 00:00:58.335 - entrepreneur 28 00:00:59.130 --> 00:01:00.350 - health care founder. 29 00:01:00.809 --> 00:01:02.829 - We're joined today by Nish Kandwala. 30 00:01:03.530 --> 00:01:05.709 - And Nish is the CEO and the cofounder 31 00:01:05.849 --> 00:01:07.229 - of Bunker Hill Health. 32 00:01:07.530 --> 00:01:10.109 - We're gonna talk about what Bunker Hill does, 33 00:01:10.729 --> 00:01:12.269 - about Nish's Stanford 34 00:01:12.569 --> 00:01:14.655 - background, and and a lot more. 35 00:01:15.055 --> 00:01:16.814 - Nish, can you take a moment and tell 36 00:01:16.814 --> 00:01:18.734 - us a little bit about yourself, a little 37 00:01:18.734 --> 00:01:20.974 - bit about Bunker Hill Health, and how you 38 00:01:20.974 --> 00:01:23.394 - got to the spot of founding this company, 39 00:01:23.854 --> 00:01:25.854 - that's been funded by some of the most 40 00:01:25.854 --> 00:01:28.275 - well known venture capital funds in the universe. 41 00:01:28.549 --> 00:01:30.230 - Talk about how you got here and about 42 00:01:30.230 --> 00:01:32.170 - yourself and about Bunker Hill Health. 43 00:01:32.630 --> 00:01:33.450 - Sounds wonderful. 44 00:01:34.549 --> 00:01:37.030 - Thanks for having me, Scott. For the audience, 45 00:01:37.030 --> 00:01:39.189 - my name is Nish. I am the CEO 46 00:01:39.189 --> 00:01:41.930 - and cofounder of Bunker Hill, as Scott mentioned. 47 00:01:42.405 --> 00:01:45.045 - I was previously a student at Stanford working 48 00:01:45.045 --> 00:01:47.784 - in the AI for medicine labs over there, 49 00:01:48.325 --> 00:01:48.825 - where 50 00:01:49.284 --> 00:01:51.444 - a lot of my work was actually focused 51 00:01:51.444 --> 00:01:54.165 - on building different AI algorithms for different use 52 00:01:54.165 --> 00:01:56.724 - cases in health care. We worked very closely 53 00:01:56.724 --> 00:01:57.465 - with physicians 54 00:01:58.120 --> 00:01:59.799 - and the school of medicine, and we would 55 00:01:59.799 --> 00:02:02.379 - build some just absolutely wonderful algorithms 56 00:02:02.760 --> 00:02:04.540 - that would drive better patient care. 57 00:02:05.000 --> 00:02:05.500 - And 58 00:02:06.280 --> 00:02:09.080 - at some point, we had this realization that 59 00:02:09.080 --> 00:02:10.939 - we had built so many algorithms 60 00:02:11.479 --> 00:02:12.699 - that seemed to work, 61 00:02:13.395 --> 00:02:15.474 - but we hadn't actually gotten any of them 62 00:02:15.474 --> 00:02:15.974 - deployed. 63 00:02:16.914 --> 00:02:19.395 - That it seemed like there was a lot 64 00:02:19.395 --> 00:02:22.215 - of buzz around AI, but when the Jabram 65 00:02:22.354 --> 00:02:24.435 - hit the road, we didn't actually see many 66 00:02:24.435 --> 00:02:26.294 - of them eventually get deployed. 67 00:02:26.750 --> 00:02:28.990 - And so we got really worried about this 68 00:02:28.990 --> 00:02:31.229 - because the whole point of this is to 69 00:02:31.229 --> 00:02:33.489 - not just publish papers, it's to actually 70 00:02:33.789 --> 00:02:36.770 - drive better patient care, drive better operational efficiencies, 71 00:02:37.150 --> 00:02:39.090 - and we were just not seeing that translation 72 00:02:39.229 --> 00:02:39.729 - happen. 73 00:02:40.275 --> 00:02:42.294 - And we also saw 74 00:02:42.835 --> 00:02:45.814 - that this problem was only gonna get worse 75 00:02:46.435 --> 00:02:48.514 - with large language models. When I was a 76 00:02:48.514 --> 00:02:50.675 - student, large language models weren't a thing, so 77 00:02:50.675 --> 00:02:52.435 - it actually took us three to six months 78 00:02:52.435 --> 00:02:53.415 - to build a model 79 00:02:54.290 --> 00:02:56.210 - for a given use case. But now with 80 00:02:56.210 --> 00:02:58.530 - large language models, you could build those models 81 00:02:58.530 --> 00:03:00.710 - even faster. You could build those use cases 82 00:03:00.770 --> 00:03:03.569 - even faster. But that problem of a lot 83 00:03:03.569 --> 00:03:06.290 - of potential but not so much actually live 84 00:03:06.290 --> 00:03:08.710 - in production use, that problem only got exacerbated. 85 00:03:09.215 --> 00:03:11.375 - So we spun out Bunker Hill with the 86 00:03:11.375 --> 00:03:12.674 - sole goal of bringing 87 00:03:13.215 --> 00:03:14.114 - AI from 88 00:03:14.414 --> 00:03:17.614 - people's might ideas and bringing them into a 89 00:03:17.614 --> 00:03:19.794 - clinical practice and into production use. 90 00:03:20.655 --> 00:03:22.974 - Since then, we are now live with over 91 00:03:22.974 --> 00:03:25.650 - 30 different health systems and have raised around 92 00:03:25.650 --> 00:03:26.689 - $60,000,000 93 00:03:26.689 --> 00:03:28.849 - from some of the best venture capitalists like 94 00:03:28.849 --> 00:03:30.790 - Sequoia Capital and Khosla Ventures. 95 00:03:31.569 --> 00:03:33.810 - Take a second, Don. Give the audience a 96 00:03:33.810 --> 00:03:36.069 - sense. You're working with 60 different health systems. 97 00:03:36.290 --> 00:03:38.210 - Mhmm. Give us a sense of what Bunker 98 00:03:38.210 --> 00:03:40.555 - Hill Health actually does. What do you folks 99 00:03:40.555 --> 00:03:42.314 - do so people could just get that much 100 00:03:42.314 --> 00:03:44.334 - of a clearer sense to that? Of course. 101 00:03:44.794 --> 00:03:46.955 - So we are an AI agent platform, and 102 00:03:46.955 --> 00:03:49.854 - it's a way for clinical and operational teams 103 00:03:49.914 --> 00:03:53.019 - within health systems to build AI agents that 104 00:03:53.019 --> 00:03:54.879 - end up doing the work end to end. 105 00:03:55.419 --> 00:03:57.739 - So today, you know, you everyone has a 106 00:03:57.739 --> 00:03:59.659 - lot of ideas around how they might want 107 00:03:59.659 --> 00:04:01.519 - to use chat g p t, for example, 108 00:04:01.579 --> 00:04:03.099 - in their day to day work within the 109 00:04:03.099 --> 00:04:06.525 - health system, whether that's for something like enabling 110 00:04:06.525 --> 00:04:09.164 - care pathways on the clinical side to something 111 00:04:09.164 --> 00:04:10.305 - like helping with, 112 00:04:10.685 --> 00:04:13.245 - triaging patients in a referral queue for patient 113 00:04:13.245 --> 00:04:15.644 - access to something on the prior auth side 114 00:04:15.644 --> 00:04:17.850 - to something on the registry data abstraction side. 115 00:04:17.930 --> 00:04:20.029 - There are so many of these use cases, 116 00:04:20.170 --> 00:04:22.250 - but we have found that health systems struggle 117 00:04:22.250 --> 00:04:25.230 - to actually get those ideas about how 118 00:04:25.689 --> 00:04:27.689 - technologies like chat g p t can enable 119 00:04:27.689 --> 00:04:30.250 - with those use cases actually live into production 120 00:04:30.250 --> 00:04:32.194 - use. So what we have built is this 121 00:04:32.194 --> 00:04:33.735 - platform. We call it Carebricks, 122 00:04:34.115 --> 00:04:36.454 - and it's a way for clinical and operational 123 00:04:36.514 --> 00:04:38.435 - teams to work with us or to build 124 00:04:38.435 --> 00:04:41.154 - them by themselves by taking our platform and 125 00:04:41.154 --> 00:04:43.235 - building AI agents that end up doing the 126 00:04:43.235 --> 00:04:45.360 - work end to end. Each of these use 127 00:04:45.360 --> 00:04:47.459 - cases that I mentioned and many more, 128 00:04:48.000 --> 00:04:50.639 - you can probably distill them down to three 129 00:04:50.639 --> 00:04:51.139 - components. 130 00:04:51.600 --> 00:04:53.519 - There is one component where you need to 131 00:04:53.519 --> 00:04:55.919 - get access to data from multiple systems of 132 00:04:55.919 --> 00:04:56.419 - record. 133 00:04:56.834 --> 00:04:57.334 - Think 134 00:04:57.795 --> 00:05:00.595 - EHR. Think ERP. Think other sources of records 135 00:05:00.595 --> 00:05:03.975 - like clinical guidelines, payer policies, things like that. 136 00:05:04.035 --> 00:05:05.014 - That's one component. 137 00:05:05.394 --> 00:05:08.115 - Another component is that of actually running the 138 00:05:08.115 --> 00:05:10.055 - large language models on top of, 139 00:05:10.620 --> 00:05:12.620 - that data, and then you have a final 140 00:05:12.620 --> 00:05:14.639 - component of actually taking actions 141 00:05:15.020 --> 00:05:16.780 - on top of the output of AI. This 142 00:05:16.780 --> 00:05:19.439 - is things like writing back to the EHR, 143 00:05:20.060 --> 00:05:23.360 - things like sending patients' messages or physicians' messages. 144 00:05:23.644 --> 00:05:24.705 - And so this platform 145 00:05:25.324 --> 00:05:25.824 - allows 146 00:05:26.205 --> 00:05:28.464 - clinical and operational leaders to arrange 147 00:05:29.165 --> 00:05:30.225 - and sort of and 148 00:05:30.605 --> 00:05:31.105 - assemble 149 00:05:31.644 --> 00:05:32.464 - these components, 150 00:05:32.764 --> 00:05:34.685 - these bricks as we call them, a bit 151 00:05:34.764 --> 00:05:36.925 - that's where the name Carebricks comes from, to 152 00:05:36.925 --> 00:05:38.144 - assemble these bricks 153 00:05:38.470 --> 00:05:40.569 - in a way that we can help automate 154 00:05:40.629 --> 00:05:43.129 - many of those clinical and operational use cases. 155 00:05:44.310 --> 00:05:46.470 - And take a moment, Nish, on how do 156 00:05:46.470 --> 00:05:48.250 - systems work with you? 157 00:05:48.629 --> 00:05:49.370 - Who internally 158 00:05:49.750 --> 00:05:51.689 - works with your platform? 159 00:05:52.814 --> 00:05:54.654 - How much help do they get from your 160 00:05:54.654 --> 00:05:55.154 - team? 161 00:05:55.615 --> 00:05:57.235 - How does this work in practice? 162 00:05:58.895 --> 00:06:00.814 - Does the team internally have to be all 163 00:06:00.814 --> 00:06:01.715 - software engineers? 164 00:06:02.415 --> 00:06:04.514 - Or or or how does this work with 165 00:06:04.669 --> 00:06:06.209 - the people at the system 166 00:06:06.589 --> 00:06:08.350 - so to make this usable for the system 167 00:06:08.350 --> 00:06:10.350 - without them having to all be computer engineers? 168 00:06:10.350 --> 00:06:11.169 - Or do they? 169 00:06:11.470 --> 00:06:13.709 - Yeah. No. That's a great question. So we 170 00:06:13.709 --> 00:06:16.269 - think that this is actually a great time 171 00:06:16.269 --> 00:06:18.529 - to make an operational change as well. So 172 00:06:18.834 --> 00:06:20.914 - we partner when we work with the health 173 00:06:20.914 --> 00:06:22.854 - system, we are typically working with, 174 00:06:23.474 --> 00:06:25.314 - someone like a chief AI officer or a 175 00:06:25.314 --> 00:06:28.274 - chief medical officer where we partner very closely 176 00:06:28.274 --> 00:06:30.914 - with them to have Carebricks become their AI 177 00:06:30.914 --> 00:06:31.414 - platform. 178 00:06:31.954 --> 00:06:34.649 - And we create this governance structure when when 179 00:06:34.649 --> 00:06:36.110 - when there are new use cases 180 00:06:36.490 --> 00:06:38.729 - that are coming from the grassroots or when 181 00:06:38.729 --> 00:06:41.129 - we introduce new use cases that we have 182 00:06:41.129 --> 00:06:43.289 - found, that receive a lot of love at 183 00:06:43.289 --> 00:06:44.269 - other health systems. 184 00:06:44.649 --> 00:06:47.470 - The chief AI officer, the chief medical officer 185 00:06:47.529 --> 00:06:49.529 - is able to do a very quick sniff 186 00:06:49.529 --> 00:06:51.425 - test and see if that is of interest 187 00:06:51.425 --> 00:06:52.324 - to their organization. 188 00:06:52.865 --> 00:06:54.625 - And if that use case does pass that 189 00:06:54.625 --> 00:06:55.444 - sniff test, 190 00:06:55.824 --> 00:06:57.524 - then the chief AI officer 191 00:06:57.824 --> 00:06:58.324 - introduces 192 00:06:59.185 --> 00:07:01.584 - us to the problem owners, those use case 193 00:07:01.584 --> 00:07:04.910 - owners at their health system. And then our 194 00:07:04.910 --> 00:07:07.810 - forward deployed team works very closely 195 00:07:08.269 --> 00:07:11.009 - with the subject matter experts, those problem owners, 196 00:07:11.069 --> 00:07:13.790 - to rapidly build AI agents that help them 197 00:07:13.790 --> 00:07:16.189 - for their particular problem. And so to give 198 00:07:16.189 --> 00:07:17.649 - you a very concrete example, 199 00:07:18.004 --> 00:07:19.925 - from the first time we meet with a 200 00:07:19.925 --> 00:07:22.164 - problem owner or subject matter expert at a 201 00:07:22.164 --> 00:07:22.664 - hospital 202 00:07:23.204 --> 00:07:25.365 - to going live, it takes about two to 203 00:07:25.365 --> 00:07:28.004 - six weeks end to end, for a net 204 00:07:28.004 --> 00:07:30.985 - new use case. And so we don't expect 205 00:07:31.044 --> 00:07:33.529 - the health system to have any engineers 206 00:07:33.990 --> 00:07:36.949 - responsible for this. We just, expect that we 207 00:07:36.949 --> 00:07:37.849 - will have a collaboration 208 00:07:38.149 --> 00:07:40.470 - where our team can rapidly build these AI 209 00:07:40.470 --> 00:07:43.050 - agents on our platform for those use cases. 210 00:07:43.829 --> 00:07:46.389 - And what does your forward deployed team look 211 00:07:46.389 --> 00:07:49.564 - like? Yeah. Because that's so critical, that intersection 212 00:07:49.704 --> 00:07:52.504 - with the with the client hospital. Because we 213 00:07:52.504 --> 00:07:54.204 - see so many people that have great technology. 214 00:07:54.745 --> 00:07:56.504 - But if you don't have that deployment team 215 00:07:56.504 --> 00:07:59.305 - that's really working hand in hand, it it's 216 00:07:59.305 --> 00:08:01.144 - almost what you talked about originally when you 217 00:08:01.144 --> 00:08:03.329 - were back in the day, a graduate student 218 00:08:03.329 --> 00:08:05.810 - at Stanford. Yeah. And people had ideas, but 219 00:08:05.810 --> 00:08:07.729 - they weren't really getting put to work. Talk 220 00:08:07.729 --> 00:08:09.810 - about that integration with the forward deployed team 221 00:08:09.810 --> 00:08:10.629 - to make this 222 00:08:11.329 --> 00:08:12.550 - as useful as possible. 223 00:08:12.930 --> 00:08:15.169 - Yeah. I think our forward deployed team is 224 00:08:15.169 --> 00:08:16.310 - incredibly curious 225 00:08:16.935 --> 00:08:18.875 - and is very willing to be collaborative. 226 00:08:19.335 --> 00:08:21.814 - And so to give you an idea, our 227 00:08:21.814 --> 00:08:23.194 - VP of forward deployed, 228 00:08:23.574 --> 00:08:25.895 - he has a background in consulting as well 229 00:08:25.895 --> 00:08:28.214 - as in start ups. He studied applied math 230 00:08:28.214 --> 00:08:29.355 - at Brown University. 231 00:08:29.939 --> 00:08:33.000 - And when he was the only forward deployed, 232 00:08:33.700 --> 00:08:34.919 - lead that we had, 233 00:08:35.379 --> 00:08:37.540 - he would, you know, fly out to the 234 00:08:37.540 --> 00:08:40.019 - health systems that we partner with, meet with 235 00:08:40.019 --> 00:08:42.884 - the stakeholders, the problem owners. And just if 236 00:08:42.884 --> 00:08:45.044 - they are already doing a process, but just 237 00:08:45.044 --> 00:08:45.544 - manually, 238 00:08:46.325 --> 00:08:48.884 - we would observe what they're doing, shadow them, 239 00:08:48.884 --> 00:08:51.365 - and try to basically mimic that workflow on 240 00:08:51.365 --> 00:08:54.504 - our platform, obviously, in a more automated way. 241 00:08:54.565 --> 00:08:55.865 - Or if it's a process 242 00:08:56.169 --> 00:08:58.250 - that they haven't the health health system hasn't 243 00:08:58.250 --> 00:09:01.129 - implemented yet but is interested in doing so, 244 00:09:01.129 --> 00:09:03.210 - then it's a long form conversation that we 245 00:09:03.210 --> 00:09:06.269 - have and then very close iteration and collaboration 246 00:09:06.410 --> 00:09:08.090 - on it as we go live with that 247 00:09:08.090 --> 00:09:08.830 - use case. 248 00:09:09.545 --> 00:09:12.205 - How much is the use of AI accelerating 249 00:09:12.345 --> 00:09:14.345 - in health systems today? What do you sort 250 00:09:14.345 --> 00:09:15.245 - of see now, 251 00:09:15.865 --> 00:09:17.705 - use cases? And over the next three to 252 00:09:17.705 --> 00:09:19.705 - five years, where you sort of seeing the 253 00:09:19.705 --> 00:09:20.605 - most interest 254 00:09:20.985 --> 00:09:22.985 - in in in how are things accelerating or 255 00:09:22.985 --> 00:09:23.485 - cascading? 256 00:09:24.649 --> 00:09:27.209 - So I think of use cases actually in 257 00:09:27.209 --> 00:09:29.370 - the form of, like, a Maslow's hierarchy of 258 00:09:29.370 --> 00:09:30.669 - needs at a health system, 259 00:09:31.449 --> 00:09:33.850 - where you have your base foundational needs that 260 00:09:33.850 --> 00:09:35.690 - need to be met. And then as you 261 00:09:35.690 --> 00:09:37.529 - go towards the top of the pyramid, you 262 00:09:37.529 --> 00:09:40.285 - hit those enlightened use cases that are very 263 00:09:40.285 --> 00:09:42.445 - cool, that are very awesome. They make a 264 00:09:42.445 --> 00:09:44.305 - lot of sense from an ROI perspective, 265 00:09:44.764 --> 00:09:47.105 - but they are not foundational in the nature. 266 00:09:47.165 --> 00:09:49.565 - So, like, an example of a foundational use 267 00:09:49.565 --> 00:09:52.705 - case would be, oh, wow. With the new 268 00:09:52.925 --> 00:09:54.144 - HR one bill, 269 00:09:54.679 --> 00:09:57.159 - the definition of who is considered a Medicaid 270 00:09:57.159 --> 00:09:58.620 - patient is about to change, 271 00:09:59.159 --> 00:10:00.539 - and our hospital 272 00:10:01.000 --> 00:10:03.659 - might now have a lower percentage of Medicaid 273 00:10:03.720 --> 00:10:04.220 - patients, 274 00:10:05.000 --> 00:10:07.159 - than what is required to be eligible for 275 00:10:07.159 --> 00:10:08.620 - three forty b discounts. 276 00:10:09.365 --> 00:10:12.245 - And suddenly, you are at risk as a 277 00:10:12.245 --> 00:10:14.804 - hospital of losing tens, if not hundreds, of 278 00:10:14.804 --> 00:10:15.945 - millions of dollars, 279 00:10:16.325 --> 00:10:18.164 - that are afforded by the three forty b 280 00:10:18.164 --> 00:10:20.825 - program. And so that is like a foundational 281 00:10:20.964 --> 00:10:23.399 - need. It is very difficult to care about 282 00:10:23.399 --> 00:10:25.179 - other things when you have such a foundational 283 00:10:25.240 --> 00:10:27.159 - need that is unmet. So that will be 284 00:10:27.159 --> 00:10:29.820 - a great use case to anchor our collaboration 285 00:10:30.039 --> 00:10:32.120 - on where you could have an AI agent, 286 00:10:32.120 --> 00:10:32.860 - for example, 287 00:10:33.720 --> 00:10:36.199 - looking at every patient that is underinsured or 288 00:10:36.199 --> 00:10:36.699 - uninsured, 289 00:10:37.264 --> 00:10:39.184 - seeing which of those patients would actually be 290 00:10:39.184 --> 00:10:40.325 - eligible for Medicaid, 291 00:10:40.705 --> 00:10:42.784 - and getting them enrolled with the help of 292 00:10:42.784 --> 00:10:43.524 - some assistance, 293 00:10:43.904 --> 00:10:46.144 - for example. So I think of these use 294 00:10:46.144 --> 00:10:48.144 - cases as these are, like, the foundational use 295 00:10:48.144 --> 00:10:50.965 - cases at the base of the Maslow's hierarchy 296 00:10:51.024 --> 00:10:53.169 - pyramid, and then you have use cases at 297 00:10:53.169 --> 00:10:55.269 - the very top that are, like, you know, 298 00:10:55.330 --> 00:10:58.450 - incredibly frontier care pathways that you have in 299 00:10:58.450 --> 00:11:00.690 - mind. Like, at the for example, at the 300 00:11:00.690 --> 00:11:02.929 - Cleveland Clinic, an example of a frontier use 301 00:11:02.929 --> 00:11:05.284 - case is one where they are doing trials 302 00:11:05.424 --> 00:11:08.804 - to understand if patients with moderate aortic stenosis 303 00:11:08.865 --> 00:11:12.565 - would benefit from from earlier intervention, for example. 304 00:11:12.625 --> 00:11:14.304 - And so they are very interested in a 305 00:11:14.304 --> 00:11:15.605 - care pathway specifically 306 00:11:16.519 --> 00:11:19.480 - for a frontier care pathway like that, and 307 00:11:19.480 --> 00:11:21.080 - that would be an AI agent that helps 308 00:11:21.080 --> 00:11:24.040 - with patient navigation and follow-up, for example. And 309 00:11:24.040 --> 00:11:25.960 - so I think of it in terms of 310 00:11:25.960 --> 00:11:28.120 - when we partner with the health system, we 311 00:11:28.120 --> 00:11:30.475 - want to get a few use cases that 312 00:11:30.475 --> 00:11:33.375 - cover the foundation, a few use cases that 313 00:11:33.434 --> 00:11:35.514 - span across that pyramid going all the way 314 00:11:35.514 --> 00:11:37.514 - up to the sort of, the top of 315 00:11:37.514 --> 00:11:39.195 - the the top of the pyramid into the 316 00:11:39.195 --> 00:11:40.415 - enlightened use cases. 317 00:11:41.915 --> 00:11:43.835 - Thank you. Absolutely fascinating. They go up the 318 00:11:43.835 --> 00:11:46.070 - Maslow's hierarchy type of perspective, breaking it down 319 00:11:46.070 --> 00:11:48.790 - to what's foundational versus what's frontier, what's absolutely 320 00:11:48.790 --> 00:11:50.790 - needed versus what's nice to have or next 321 00:11:50.790 --> 00:11:53.549 - level. Exactly. Talk a little bit about what 322 00:11:53.549 --> 00:11:55.690 - are you most focused on and excited about 323 00:11:56.389 --> 00:11:57.769 - this next coming year 324 00:11:58.164 --> 00:12:00.565 - both in in your work with your health 325 00:12:00.565 --> 00:12:02.485 - systems, your work with your team, your work 326 00:12:02.485 --> 00:12:03.225 - with your 327 00:12:03.924 --> 00:12:05.845 - venture capital funds? What what are you most 328 00:12:05.845 --> 00:12:07.524 - excited about and focused on as you as 329 00:12:07.524 --> 00:12:08.965 - you work for the next six to twelve 330 00:12:08.965 --> 00:12:09.465 - months? 331 00:12:10.245 --> 00:12:12.745 - I think it's two things. The first one 332 00:12:13.299 --> 00:12:15.220 - is actually something that I hope we can 333 00:12:15.220 --> 00:12:16.440 - develop as an ecosystem 334 00:12:16.899 --> 00:12:17.639 - as a whole, 335 00:12:18.259 --> 00:12:20.519 - and that is how do you actually 336 00:12:21.779 --> 00:12:23.379 - look at AI as an a as a 337 00:12:23.460 --> 00:12:25.779 - as an enterprise strategy? What I mean by 338 00:12:25.779 --> 00:12:28.085 - this is, you know, as we think about 339 00:12:28.144 --> 00:12:30.144 - how we even want to make a business 340 00:12:30.144 --> 00:12:31.044 - model around 341 00:12:31.504 --> 00:12:32.245 - our platform, 342 00:12:32.705 --> 00:12:35.184 - you know, one approach would have been to 343 00:12:35.184 --> 00:12:37.764 - sort of charge use case by use case. 344 00:12:38.144 --> 00:12:40.410 - But I can tell you that it creates 345 00:12:40.410 --> 00:12:42.649 - a lot of friction when every time there's 346 00:12:42.649 --> 00:12:44.110 - a new use case in mind, 347 00:12:44.410 --> 00:12:46.009 - you have to go back to the legal 348 00:12:46.009 --> 00:12:48.809 - team, negotiate a price point, get that down, 349 00:12:48.809 --> 00:12:50.509 - and that sort of creates a, 350 00:12:51.049 --> 00:12:53.790 - you know, loss of momentum over several months 351 00:12:54.024 --> 00:12:55.784 - as an example. So that's, like, you know, 352 00:12:55.784 --> 00:12:57.945 - a business model that is not really what 353 00:12:57.945 --> 00:13:00.284 - we really appreciate because it really reduces 354 00:13:00.825 --> 00:13:02.664 - the the sort of how much you can 355 00:13:02.664 --> 00:13:04.985 - use AI for different use cases. So where 356 00:13:04.985 --> 00:13:06.985 - we are converging towards and we were where 357 00:13:06.985 --> 00:13:09.404 - we would love to develop this ecosystem towards 358 00:13:09.779 --> 00:13:12.419 - is thinking of AI similar to how you 359 00:13:12.419 --> 00:13:13.320 - think of electricity. 360 00:13:13.860 --> 00:13:15.639 - Just like when you use electricity, 361 00:13:16.419 --> 00:13:17.639 - you pay for watts, 362 00:13:18.100 --> 00:13:20.820 - for every unit of consumption, in a similar 363 00:13:20.820 --> 00:13:22.360 - way, we hope that 364 00:13:22.664 --> 00:13:25.544 - the ecosystem start starts pricing AI similar to 365 00:13:25.544 --> 00:13:26.044 - electricity 366 00:13:26.504 --> 00:13:28.524 - where you pay for tokens of consumption 367 00:13:28.904 --> 00:13:29.804 - instead of, 368 00:13:30.264 --> 00:13:32.205 - instead of on a per use case basis. 369 00:13:32.745 --> 00:13:35.404 - What that will result in is this rapid 370 00:13:35.709 --> 00:13:38.029 - explosion of usage of AI within a health 371 00:13:38.029 --> 00:13:40.589 - system where you can go from, you know, 372 00:13:40.589 --> 00:13:43.549 - zero to twenty, thirty plus use cases without 373 00:13:43.549 --> 00:13:46.029 - having to constantly think about, oh, wow. We 374 00:13:46.029 --> 00:13:47.470 - need to go back to the legal team 375 00:13:47.470 --> 00:13:49.514 - again and again. So that's the first thing 376 00:13:49.514 --> 00:13:51.195 - that I'm really focused on, which is how 377 00:13:51.195 --> 00:13:54.095 - can we shape this ecosystem to view AI 378 00:13:54.235 --> 00:13:57.434 - similar to electricity as opposed to a host 379 00:13:57.434 --> 00:13:58.414 - of point solutions. 380 00:13:59.195 --> 00:14:02.509 - That's number one. And number two is actually 381 00:14:02.509 --> 00:14:05.629 - around this the speed at which AI itself 382 00:14:05.629 --> 00:14:08.990 - is improving and what capabilities it affords actually 383 00:14:08.990 --> 00:14:11.309 - unlocks a whole host of use cases that 384 00:14:11.309 --> 00:14:13.710 - previously were just not possible. So what I 385 00:14:13.710 --> 00:14:16.210 - am focused on internally within the company is 386 00:14:16.565 --> 00:14:18.584 - over the course of the last six months, 387 00:14:18.804 --> 00:14:21.144 - AI has progressed so much with new technologies 388 00:14:21.445 --> 00:14:24.004 - or new capabilities like computer use. How do 389 00:14:24.004 --> 00:14:27.524 - we capture those advanced capabilities in unlocking even 390 00:14:27.524 --> 00:14:28.504 - more use cases 391 00:14:28.899 --> 00:14:31.059 - with our for our partners and unlocking even 392 00:14:31.059 --> 00:14:33.860 - more value creation with them. And so those 393 00:14:33.860 --> 00:14:35.620 - are the two things that I'm really focused 394 00:14:35.620 --> 00:14:38.039 - on. One is viewing AI's electricity, 395 00:14:38.419 --> 00:14:40.100 - and two is how do we keep up 396 00:14:40.100 --> 00:14:41.879 - with the pace of the broader, 397 00:14:42.339 --> 00:14:44.600 - inherent AI development that's going on. 398 00:14:45.595 --> 00:14:48.154 - You've had a chance over the last few 399 00:14:48.154 --> 00:14:50.235 - years to work with some of the highest 400 00:14:50.235 --> 00:14:52.735 - end venture capital names in the world, 401 00:14:53.115 --> 00:14:53.615 - Sequoia, 402 00:14:54.154 --> 00:14:56.014 - Khosla Ventures, and so forth. 403 00:14:56.795 --> 00:14:59.910 - Give people just some quick insight as to 404 00:14:59.910 --> 00:15:02.250 - what it's like to work with those 405 00:15:02.870 --> 00:15:05.429 - funding sources, those people. What is it like? 406 00:15:05.429 --> 00:15:07.450 - I mean, you're dealing with really bright people. 407 00:15:07.590 --> 00:15:09.210 - Yes. How do you connect with them? 408 00:15:10.084 --> 00:15:11.784 - Any thoughts or advice about 409 00:15:12.164 --> 00:15:15.365 - people working with funds like that? Yeah. No. 410 00:15:15.365 --> 00:15:17.544 - No. That's a that's a great point. So 411 00:15:17.684 --> 00:15:20.164 - even at Sequoia, we work with Alfred Lin, 412 00:15:20.164 --> 00:15:21.704 - who is the head of the Sequoia 413 00:15:22.164 --> 00:15:24.470 - team. And at and at Khosla, we work 414 00:15:24.470 --> 00:15:26.570 - with Vinod Khosla himself. He is, 415 00:15:27.029 --> 00:15:27.529 - obviously, 416 00:15:28.870 --> 00:15:31.050 - an entrepreneur of of our generation. 417 00:15:31.910 --> 00:15:32.410 - And 418 00:15:33.110 --> 00:15:35.190 - so what I'm the reason I shared this 419 00:15:35.190 --> 00:15:35.850 - is because 420 00:15:36.254 --> 00:15:38.254 - those two guys, they have nothing left to 421 00:15:38.254 --> 00:15:38.754 - prove. 422 00:15:39.134 --> 00:15:41.615 - So when they think about investing in the 423 00:15:41.615 --> 00:15:43.235 - future, investing in companies, 424 00:15:43.535 --> 00:15:45.295 - they are not trying to think of, oh, 425 00:15:45.295 --> 00:15:47.215 - how can I make a quick buck? They 426 00:15:47.215 --> 00:15:49.054 - are thinking about this in the sense of 427 00:15:49.054 --> 00:15:51.769 - how can we invest in ideas, in companies, 428 00:15:51.769 --> 00:15:53.230 - in founders, in teams 429 00:15:53.529 --> 00:15:55.289 - that can really change the shape of a 430 00:15:55.289 --> 00:15:56.429 - ginormous industry. 431 00:15:57.210 --> 00:15:58.889 - And that's how they are looking at us. 432 00:15:58.889 --> 00:16:01.450 - So when they invested in us, when we 433 00:16:01.450 --> 00:16:03.129 - were chatting with them and we continue to 434 00:16:03.129 --> 00:16:03.950 - work with them, 435 00:16:04.565 --> 00:16:07.544 - where they were coming from is, look, today, 436 00:16:07.605 --> 00:16:09.764 - health care is usually criticized to be the 437 00:16:09.764 --> 00:16:11.865 - last industry to adopt new technologies, 438 00:16:12.485 --> 00:16:15.125 - and we have always criticized health care to 439 00:16:15.125 --> 00:16:17.524 - be one where things are not nimble, things 440 00:16:17.524 --> 00:16:19.379 - take a lot of time. And how they 441 00:16:19.379 --> 00:16:21.620 - see Bunker Hill, how they see our platform 442 00:16:21.620 --> 00:16:22.120 - Carebricks, 443 00:16:22.500 --> 00:16:24.500 - is actually a way to make health care 444 00:16:24.500 --> 00:16:25.319 - more iterable. 445 00:16:25.940 --> 00:16:27.620 - How can you make it such that the 446 00:16:27.620 --> 00:16:28.600 - cost of iteration 447 00:16:29.139 --> 00:16:31.860 - reduces down close to zero as possible? How 448 00:16:31.860 --> 00:16:33.299 - can you make it such that the speed 449 00:16:33.299 --> 00:16:33.959 - of iteration, 450 00:16:34.855 --> 00:16:37.514 - in health care increases as much as possible 451 00:16:37.815 --> 00:16:40.075 - so that when you have a new idea, 452 00:16:40.295 --> 00:16:42.215 - it does not take half a decade or 453 00:16:42.215 --> 00:16:44.695 - a decade to get that into widespread use, 454 00:16:44.695 --> 00:16:46.875 - that you can make it a lot more, 455 00:16:47.290 --> 00:16:48.970 - sooner and you can do it in a 456 00:16:48.970 --> 00:16:51.769 - way that feels very nimble and that feels 457 00:16:51.769 --> 00:16:52.269 - very, 458 00:16:52.649 --> 00:16:55.129 - very fast and safe. And so they think 459 00:16:55.129 --> 00:16:56.889 - of that as if Bunker Hill can do 460 00:16:56.889 --> 00:16:59.529 - that, it unlocks a whole host of value 461 00:16:59.529 --> 00:17:01.610 - creation for the entire world. And so that's 462 00:17:01.610 --> 00:17:04.065 - why they started working with us and we 463 00:17:04.065 --> 00:17:06.244 - continue to work very closely with them. 464 00:17:07.184 --> 00:17:08.884 - And what you know, unlike 465 00:17:09.265 --> 00:17:11.984 - potentially other investors that that we that we 466 00:17:11.984 --> 00:17:12.484 - know, 467 00:17:12.945 --> 00:17:15.265 - where they encourage us to to do is 468 00:17:15.424 --> 00:17:16.785 - what they encourage us to do is to 469 00:17:16.785 --> 00:17:20.299 - actually swing big, which is don't make compromises 470 00:17:20.599 --> 00:17:23.240 - just so that you can have a a 471 00:17:23.240 --> 00:17:25.320 - quick win or, you know, a short term 472 00:17:25.320 --> 00:17:27.559 - win as opposed to compromising for the larger 473 00:17:27.559 --> 00:17:30.140 - vision. Take bigger risks. Go bigger. 474 00:17:30.525 --> 00:17:32.125 - We are here, you know, we are here 475 00:17:32.125 --> 00:17:34.605 - to support your grand vision. And so I 476 00:17:34.605 --> 00:17:36.924 - think I feel very privileged in the sense 477 00:17:36.924 --> 00:17:40.125 - that these investors are these partners we know 478 00:17:40.125 --> 00:17:41.965 - than Alfred are really able to push us 479 00:17:41.965 --> 00:17:44.420 - to think big. And, of course, you know, 480 00:17:44.420 --> 00:17:46.820 - they just come with tons of experience about 481 00:17:46.820 --> 00:17:47.559 - how to build 482 00:17:48.180 --> 00:17:51.619 - enormous companies. Like, Sequoia obviously invested in Apple, 483 00:17:51.619 --> 00:17:54.660 - Google, Airbnb, DoorDash, and Khosla was one of 484 00:17:54.660 --> 00:17:56.200 - the first investors in OpenAI. 485 00:17:56.704 --> 00:17:58.964 - They have seen how big companies get built, 486 00:17:59.105 --> 00:18:00.944 - and and they they really help us make 487 00:18:00.944 --> 00:18:03.105 - sure that we can learn from how those 488 00:18:03.105 --> 00:18:04.244 - companies got built. 489 00:18:05.184 --> 00:18:07.025 - That's fascinating. I mean, it's it's so many 490 00:18:07.025 --> 00:18:08.545 - people at the end of the day, I 491 00:18:08.545 --> 00:18:10.464 - count myself as one of them, are more 492 00:18:10.464 --> 00:18:10.964 - incrementalist 493 00:18:11.659 --> 00:18:14.220 - than thinking big people. Mhmm. How does that 494 00:18:14.220 --> 00:18:16.880 - translate from I mean, you're you're obviously 495 00:18:17.339 --> 00:18:18.159 - brilliant person, 496 00:18:18.619 --> 00:18:21.119 - straight a's at Stanford in a graduate program, 497 00:18:21.259 --> 00:18:23.200 - you know, double Stanford graduate. 498 00:18:23.659 --> 00:18:25.659 - How do you train your mind and work 499 00:18:25.659 --> 00:18:27.440 - with people to think big 500 00:18:27.875 --> 00:18:30.115 - versus looking at the next thing? How do 501 00:18:30.115 --> 00:18:31.555 - you how do you sort of combine those 502 00:18:31.555 --> 00:18:32.295 - two thoughts, 503 00:18:32.835 --> 00:18:35.715 - short and big, small and long term, big 504 00:18:35.715 --> 00:18:36.215 - thinking? 505 00:18:36.595 --> 00:18:38.434 - How how do you sort of do that 506 00:18:38.434 --> 00:18:40.134 - as a a founder, entrepreneur, 507 00:18:40.674 --> 00:18:41.654 - CEO, leader? 508 00:18:42.519 --> 00:18:44.119 - How do you evolve to do that to 509 00:18:44.119 --> 00:18:45.720 - do that well? Is is that something that 510 00:18:45.720 --> 00:18:48.200 - you evolve into or that naturally you see, 511 00:18:48.200 --> 00:18:49.799 - or how do you look at that? That 512 00:18:49.799 --> 00:18:51.339 - is such a great question. 513 00:18:51.960 --> 00:18:53.740 - You know, I I think that 514 00:18:54.974 --> 00:18:57.055 - we see this in our ecosystem too. I 515 00:18:57.055 --> 00:18:59.375 - think there's a school of thought where it's 516 00:18:59.375 --> 00:19:00.595 - like AI has 517 00:19:01.055 --> 00:19:02.194 - enormous potential 518 00:19:02.734 --> 00:19:03.714 - for health care, 519 00:19:04.734 --> 00:19:06.414 - but it also comes with a lot of 520 00:19:06.414 --> 00:19:06.914 - risks. 521 00:19:07.599 --> 00:19:09.619 - So we gotta get everything right 522 00:19:10.240 --> 00:19:12.319 - on the first try, and we gotta make 523 00:19:12.319 --> 00:19:13.839 - it such that the first time we do 524 00:19:13.839 --> 00:19:16.879 - it, it's correct and we can execute on 525 00:19:16.879 --> 00:19:19.519 - that. I think that that's a recipe for 526 00:19:19.519 --> 00:19:20.659 - failure, especially 527 00:19:21.484 --> 00:19:23.404 - when this is such a new industry and 528 00:19:23.404 --> 00:19:24.865 - it's such a fast moving industry. 529 00:19:25.565 --> 00:19:27.565 - Like, none of us here, no one in 530 00:19:27.565 --> 00:19:28.305 - this world 531 00:19:28.924 --> 00:19:31.805 - can claim that, oh, we have five years 532 00:19:31.805 --> 00:19:33.424 - or ten years of experience 533 00:19:33.884 --> 00:19:35.025 - deploying AI 534 00:19:35.329 --> 00:19:37.730 - because that AI in its current form did 535 00:19:37.730 --> 00:19:38.630 - not even exist 536 00:19:39.009 --> 00:19:40.849 - more than a couple of years ago. And 537 00:19:40.849 --> 00:19:43.809 - so I think it's important to almost take 538 00:19:43.809 --> 00:19:45.670 - a step back and be self aware 539 00:19:46.210 --> 00:19:48.289 - that we are not gonna get everything right 540 00:19:48.289 --> 00:19:49.109 - as an industry 541 00:19:49.490 --> 00:19:50.470 - on day one 542 00:19:50.794 --> 00:19:51.294 - despite 543 00:19:51.595 --> 00:19:54.174 - the potential for AI being so huge. 544 00:19:54.714 --> 00:19:57.454 - And when you have that sort of awareness, 545 00:19:58.234 --> 00:20:01.115 - what that almost translates to is that you've 546 00:20:01.115 --> 00:20:03.375 - gotta set up yourself in a way 547 00:20:04.259 --> 00:20:06.900 - where not everything feels like a one way 548 00:20:06.900 --> 00:20:07.400 - door. 549 00:20:07.859 --> 00:20:09.859 - Not everything feels like a three to five 550 00:20:09.859 --> 00:20:12.500 - year commitment when you know it's risky. Not 551 00:20:12.500 --> 00:20:15.539 - everything feels like a huge gaping hole in 552 00:20:15.539 --> 00:20:16.599 - your p and l. 553 00:20:16.994 --> 00:20:17.654 - It should 554 00:20:18.115 --> 00:20:18.855 - feel like 555 00:20:19.154 --> 00:20:21.414 - the stakes are low. It should feel 556 00:20:21.875 --> 00:20:24.535 - like it's the cost of iteration is lower. 557 00:20:24.755 --> 00:20:25.494 - And so 558 00:20:25.875 --> 00:20:27.795 - where I like to sort of think about 559 00:20:27.795 --> 00:20:29.954 - this, not just in terms of our platform, 560 00:20:29.954 --> 00:20:31.015 - but also internally, 561 00:20:31.509 --> 00:20:33.109 - is how can we make the cost of 562 00:20:33.109 --> 00:20:35.369 - making mistakes as little as possible? 563 00:20:35.910 --> 00:20:37.769 - How can we make the cost of iteration 564 00:20:37.829 --> 00:20:39.910 - as little as possible? How can we increase 565 00:20:39.910 --> 00:20:40.970 - the speed of iteration? 566 00:20:41.349 --> 00:20:42.730 - And when you have that, 567 00:20:43.349 --> 00:20:46.015 - what that results in is that you make 568 00:20:46.015 --> 00:20:47.315 - very quick strides 569 00:20:47.855 --> 00:20:49.474 - towards your ultimate vision. 570 00:20:49.855 --> 00:20:52.575 - I think you'd you'd dream big, but then 571 00:20:52.575 --> 00:20:54.815 - you know where you are today as an 572 00:20:54.815 --> 00:20:57.555 - industry, and then you have a actual path 573 00:20:57.750 --> 00:20:59.750 - that you can chart from where you currently 574 00:20:59.750 --> 00:21:01.990 - are to where you want to be and 575 00:21:01.990 --> 00:21:03.910 - where you see the potential being. And you 576 00:21:03.910 --> 00:21:05.750 - make it such that you can take small 577 00:21:05.750 --> 00:21:06.890 - steps quickly, 578 00:21:07.589 --> 00:21:08.089 - iterate, 579 00:21:08.549 --> 00:21:10.890 - backtrack if it doesn't work, and continue 580 00:21:11.269 --> 00:21:12.089 - to move towards 581 00:21:12.494 --> 00:21:14.734 - that end state. And so I think it 582 00:21:14.734 --> 00:21:17.295 - ultimately just boils down to how can we 583 00:21:17.295 --> 00:21:18.974 - make it such that you can iterate as 584 00:21:18.974 --> 00:21:20.194 - rapidly as possible. 585 00:21:21.134 --> 00:21:23.154 - No. And I and I love that perspective 586 00:21:23.214 --> 00:21:25.055 - on thinking. So I'm I'm very much an 587 00:21:25.055 --> 00:21:25.555 - incrementalist, 588 00:21:26.095 --> 00:21:27.634 - an iterative type person, 589 00:21:28.410 --> 00:21:29.769 - and and and, like, I you know, if 590 00:21:29.769 --> 00:21:31.690 - I'm writing something, it's gonna go through 10 591 00:21:31.690 --> 00:21:32.809 - drafts till I get it to where I 592 00:21:32.809 --> 00:21:34.970 - want it to be. And one of the 593 00:21:34.970 --> 00:21:36.730 - worst things that a hiring manager could do 594 00:21:36.730 --> 00:21:38.670 - at one of the companies I work with 595 00:21:38.970 --> 00:21:40.809 - is get scared of hiring because they've not 596 00:21:40.809 --> 00:21:41.710 - hit it perfectly. 597 00:21:42.464 --> 00:21:44.224 - Because, like, I don't want people to get 598 00:21:44.224 --> 00:21:45.904 - gun shy because they missed on a hire, 599 00:21:45.904 --> 00:21:48.065 - but they don't hire the next hire. That 600 00:21:48.065 --> 00:21:49.424 - life and almost everything we do is a 601 00:21:49.424 --> 00:21:51.105 - work in process. I mean, one of the 602 00:21:51.105 --> 00:21:52.625 - things you point to is the epics of 603 00:21:52.625 --> 00:21:54.464 - the world, the apples of the world, and 604 00:21:54.464 --> 00:21:56.404 - one of the things they've gotten very right 605 00:21:56.750 --> 00:21:59.869 - is its constant improvement. It it is minimum 606 00:21:59.869 --> 00:22:01.490 - viable and then constant improvement. 607 00:22:01.789 --> 00:22:03.009 - And what was Apple 608 00:22:03.710 --> 00:22:06.029 - fifteen years ago is so different than their 609 00:22:06.029 --> 00:22:08.029 - operating system today. And the same thing with 610 00:22:08.029 --> 00:22:10.210 - Epic, it was very much an an incremental 611 00:22:11.204 --> 00:22:14.825 - incremental type of effort going big, but realizing 612 00:22:14.964 --> 00:22:16.724 - it's a work in process constantly and it 613 00:22:16.724 --> 00:22:18.804 - never stops. And that seems to mirror some 614 00:22:18.804 --> 00:22:20.105 - of the thinking that you have. 615 00:22:20.644 --> 00:22:22.565 - Yeah. You'd look at the first iPhone that 616 00:22:22.565 --> 00:22:23.279 - was released 617 00:22:23.840 --> 00:22:26.259 - to the 17 Pro Max right now, 618 00:22:26.559 --> 00:22:28.480 - and you see that there's just been, you 619 00:22:28.480 --> 00:22:30.740 - know, decades of innovation that has been, 620 00:22:31.920 --> 00:22:33.840 - that has been born through through its cycles 621 00:22:33.840 --> 00:22:34.580 - of iteration. 622 00:22:35.875 --> 00:22:36.694 - Really fascinating. 623 00:22:36.994 --> 00:22:39.234 - Nish, I wanna thank you for joining us. 624 00:22:39.234 --> 00:22:40.375 - Again, Nish Kanwala, 625 00:22:41.154 --> 00:22:43.654 - CEO, cofounder of Bunker Hill Health, 626 00:22:44.115 --> 00:22:45.954 - one of the smartest people that I get 627 00:22:45.954 --> 00:22:48.110 - to visit with and talk to. What a 628 00:22:48.110 --> 00:22:49.950 - pleasure to visit with you, Nish, and I 629 00:22:49.950 --> 00:22:51.730 - look forward to visiting with you some more, 630 00:22:51.950 --> 00:22:53.869 - and and continued good luck in your journey 631 00:22:53.869 --> 00:22:55.950 - and what you're doing. Just fantastic. Thank you 632 00:22:55.950 --> 00:22:57.490 - so much, Scott. Really nice.