Audioburst Search

S2 E13: Adrienne Bolger, BlocHealth

Automatic TRANSCRIPT

Code Story is mixed and mastered in one. Click WITH CLIP game. Make Your podcasts on loud clear and clean and only one click with clip gain dot. Io being started about just finding a business model. That's what you're doing. And I think that as a summer company your roadmap needs to be as closely linked to that business model as possible. You really can't necessarily separate the business model. You're trying to find a start up right. How are you going to get paid from the features? You need that. People will pay for my name is Adrian. Bolger and I'm the founder and CTO at block health. This is code story. A podcast bring you interviews with tech visionaries who share in the critical moments. What it takes to change in industry and build an lead a team. That has your back. I'm your host no allow part. And today how Adrian Bolger set out to solve the bureaucracy in credentialing healthcare professionals using the blockchain. All this and more on Code Story Adrian. Bolger grew up in the Midwest and quickly moved up to Boston to attend him. It she's a lover of running and was a jujitsu martial arts coach at the institute growing up. She loved art and math and found her way into computer science. Trying to find the intersection of both Bolger Intern. Pixar and tinkered in robotics and medical devices and eventually she founded block health which aims to solve healthcare credentialing. End the pains associated with moving yourself your practice or even working in another state. I sat down with jared. My co-founder what is it going to take to solve this problem? Well one of the problems and credentialing is there's a lot of duplication of data silos and no one trust other people's data so we decided to use the technology from Syria octane. Open Source Distributed Ledger. To deal with the fact that no one trust each other and no one trusts where files came from a problem known as primary source for vacation in this space so that was the first thing we knew our. Mvp needed the second thing that we knew we needed was more less exotic. I would say a common application. Credentialing has its taxes way where there's so many forums. They all need the same information there. Dozens to hundreds of pages long and eat hospitals or insurance company that a health care professional wants to work with as a fifty page application. That's just slightly different from you. Know the next company down the road and so we knew that solving that problem was going to be huge for our MVP and then the third thing we knew we needed was a way of getting the data from I two pieces into out of our solution because building up you know just one. Big Data entry pile doesn't make customers very happy right. They're putting all this data and they can't get it out and the whole point of fixing this problem and smoothing the way and preventing data silos. Is that the data has to come out at some point. So those three guess you could call them sort of customer slash technical requirement. Always at the front of our minds sooner trying to figure out what we needed to build the initial version of that technology hook a bow three to four months to get off the ground and an pieces. We split it into three different parts. We had one part that we hired a contractor for that was the for blockchain technology. We license commercial software and then we also licensed the paid open source software. The software is open source but we wanted the premium support and much of the work was then sent taking those three pylons and doing the integration code on top of them the sort of keyword searches we ended up using docker We use the neen stack for our our front-end witches and back enriches Mongo expressed angular node. We use energy and exteriors us a great open source. Premiums software called Forum DOT IO COMMERCIAL PDF management software called PDF tron and then for a theory and their blockchain. The biggest tool chains that are the most supported are a library called truffle dot jazz and a tool chain called infeuro so that was a lot of software to dump on a problem but we did decide. We needed all of it right out the gate in general I think. Mvp smaller than ours wasn't but that's not what we needed. So that's not what we that's awesome so I get the blockchain technologies that you were using Gets the tools you're using there. But what made you decide to go with the mean stack over something else? Maybe some react based stuff for anything like that so honestly. Initially I had less experience an Euler than I did with react. I had worked with both. I didn't have a strong opinion. Given it we did have the luxury of starting completely greenfield and honestly brought it down on. The side of angular was the blockchain contractors that we were working with who we had already chosen and vetted for their watching expertise. Didn't have the staff to also create a front end in react. They had experts in blockchain and they had experts in angular and ultimately we decided that question expertise with more important and if we needed to switch the front end framework. Got A later date. We would just have to do that. I wish I could say that there was a great technical. You know debate between the two which is something that had actually already happened that a prior company is that to tire companies. Actually that I was at has the A degree reactors today healer debate but for the startup that it ended up not being technical reason at all as rebuilding that early version. What decisions in trade offs did you have to make in the short term? I'm leaning towards more technical debt. How did you cope with those decisions and decide okay? This is the way we're going to build it for now so that we can get out a little bit quicker. Even though you had a lot of technology you were implementing and Juggling. So some of that. Was You know the contractors where we did choose to outsource overseas for the contractors and so the iteration cycle there was very long and so pieces would come back. You know the next day two days later and deciding oh well this isn't exactly how we wanted. This style is wrong. This is wrong. This flow isn't how it was but it does work. Is it worth the risk of the contractors? Were very helpful. They would happily spend more time on it right if we wanted something slightly different sure. They're good yeah. They're contractors. That are that are paid by the hour. Project tend to be very flexible in that manner. So you have to decide at some point okay. You know. This doesn't have Google's file or drought boxes file interface. But we have to use it as a you know. This doesn't have auto show features that we need it for that. We will want to have that will feed data entry by five minutes but we need to stop here because if we don't then you're going to add another four to five days per feature by the time you implement test and get back to us and so a lot of it was the. I think some of this is both a blessing and the curse of working with some nontechnical co-founders. The standard is all of Google. The Standard is linked in wider network. Like being linked in well. Because there's only one of me says that negotiation was was definitely a key sort of peace in our decision making and then there are also some decisions on the technology side. Where you know. Initially I was like no. We're going to have automated. Continuous integration. Right from the beginning we're going to be able to scale super fast and you know a week or two into that investigation said you know. I'm looking at what you really need to start the business. Got a small Pinot setup. Single server do we really need to deploy coober? Nettie right now for our you know. Mvp and no no. You don't like it's awesome but you don't. You shouldn't spend your time on that. You should let a devops experts from their time on that and higher than three years from now. There's this mel everything's gotta be perfect. This a greenfield project and so pulling dot from the perfectionism. Is He actually getting something out? The door absolutely ready fire. Aim kind of thing. Get it out the door. You don't need coober netease at the beginning. I like that. There should probably be some sort of tagline created by all founding. Cto's that you don't need coober net. He's out of gate but I mean of course. The part of the problem was this last March and left March. Google there big coot relaunching to Rene to the world that are developer conference. That was all anyone talked about for for a solid eight weeks there. The the tree patine well actually spending alive advertising dollars on this. It's not you know. Come OUT OF NOWHERE. You guys right. We're not just playing with tools so then as you you made those decisions you built the. Mvp He you know you decided where to scale and we're not to appropriately. How did you progress the product further from that point? How did you mature it as the company grew so we got our first live customer on the product sometime around October because it was our first customer we obviously sort of went out of our way above and beyond to do as much of the initial sort of manual data entry work on the platform as possible so that we could do the first users. We could catch a lot of the problems that we were gonNA run into with this brand new product and it quickly became apparent that one thing that had been thought of as a nice to have turned out to be crucial and that was bulk. Data workflows both uploads downloads will create role based access control group folk access. Control things that are you know if you have a version that worked in singular. It's not that difficult. You think to make it all work in both. But we didn't have that and so that became the immediate priority interest bug fixing obviously for the next version. Because we realize Oh. We can't do this again. Like we literally will not be able to on board another customer if we don't have spoke import features like nope. This will not continue. This cannot continue. This episode of story is sponsored by Trista. Trista is a mobile APP. That lets you do business calling and texting from anywhere with Trista. You consider your business phone number. Download the APP and start calling and texting unlimited right away trust. It is the best business phone APP on the market. Whether you're a founder freelancers starting your business or you're already established growing your network and your business is all about communication. You've got to be available no matter where you are trust to offers the call management features that empower you to communicate smarter and more efficiently like auto attendants call recording user groups and more. And you don't need any special equipment just the smartphone. You're already using. Trista is easy to configure. So you can set everything up yourself. All online is just fifteen dollars per user per month with no contract so start your free thirty day trial today at Trista Dot com slash code story. That's www dot trista dot com slash coach story. All one word so one of my next questions is sort of how you build your roadmap and that ties in pretty nicely with what you're saying so you figure it out. You needed bulk import and a lot of different ways. You need needed a way to process some sort of mass chunk of transactions. Somehow and you figure that out some ways. What was the process for figuring out? That was the most important thing to build next and then even building on that. How do you build your roadmap so the roadmap that Lee said at West Health Right now very close very very closely tied with Your Business? Model is a quote from someone one of those famous startup books and can't remember which one which is about you know being assertive is about finding that what you're doing and I think that as a software company your roadmap needs to be as closely linked to that business model as you really can't necessarily separate the business model you're trying to find. Start up right. How were you going to get paid from features? You need that. People will pay for their very closely. So the way that we have sort of built out our roadmap is okay. We actually start with the business. Goals because my co-founder comes from sales. He's wildly optimistic. Those business schools are often not something I can build with the current team size. So then you have to go through this sort of pairing Perry back of okay. Well what software would it take to bill to get to that amount of revenue right that people want? And what can repair down? What's the biggest chunk that needs to be in place and then paired with that obviously because off where there's also just the set of things that aren't negotiable? We're moving into a world of people and small businesses and tech companies getting prosecuted for Lazy Day. Diminish it for not protecting their customer data. So one thing we've definitely spent just money on with commercial platform to go with. Our software is backups of customer data security commercial database hosting audit logging. There's a set of things that are that started to use is table stakes right. These things must be available before you can even go to her table. So you have your table stakes and then you have what you need to make money. And that is usually or then a year and a half worth of product gold and then you just prioritize by. What is the next problem? You're going. It's not a very elegant process. But I do think that it has worked really well for us because it keeps the non engineering team in the loop and it keeps it very tightly coupled which we can do because the multi right that makes sense. We'll speak into that. So that's a great segue. How did you build your team? And aligning it with product development so specifically talking about your engineering team so for Engineering Product Development. It's really important that you can find people who have eaten knowledge in the specific technology that you're already working at the stage. We are at now. We personally use as we can't afford insurance. I think that interns and new hires are sort of owed training. They're owed you know Best Practices and mentor shifts. And I think that every new engineer you looking for those things in their first job I also self aware enough to know that at the current time providing those is not going to happen so really looking for people who are very comfortable working on their own very comfortable working from written fixation and more experienced is how we've structured the wrecks for the team that were looking for and people also familiar with particular technology. But we're already working in because the cost I think of being sort of startup founder. Cto type person is that I am never going to be as good at angular at react at express at any of these technologies as a developer that Spencer so acknowledging that and just no. I'm going to find someone who is better than me at this. They're going to take. It has been a key piece of what I look for. That's awesome I definitely hear that at variable Surrounding myself with people. That are way better than me and I'm very comfortable with that. I'd like that situation so you mentioned you know not building the software when it's an MVP to be you know Uber Scalable. But how did you move towards scale ability as the product started to grow and as your team started to grow as there been challenges with scale ability? Have you been fighting it or has it sort of been a seamless transition? Tell me about that so for us for the space wherein on we have a little bit of a luxury which is that we captured one hundred percent of the credentialing market of the United States. He eight million house. So that's a lot. We're not going to get there but it also means that with the technologies that are out there in the five to seven years. Google cloud things like that And also given the fact that we aren't processing video or any large files. We are not going to hit a scale ability problem in ninety percent of our software until we're enormously successful you could handle thousands of users at the moment and it wouldn't be a problem no one's got you know gigabytes data so the scale ability issue that has kept me up as the CTO of black health is very limited. Be blockchain piece. The really looking into the specific parts of our technology that might not scale and the blockchain piece. Absolutely be it and so you're looking into standard looking at how other people already handle this problem and really looking at some open standards that are designed to deal with. This problem has been tackling it. I had a former colleague of mine. Who use this as a an interview question for interviewing engineer and he would day you if your product became an overnight success what would break and I always liked freezing it that way because it's not the the interest never nothing right. There's something that is always the limit to what you are trying to do. And you just need to know what is your system is the bottleneck and so for us at the moment. That is the blockchain technology but not only is the open source community sort of solving that for us in some ways. The standards that we've looked at we'll keep our issues with skill ability. You shouldn't be a problem until we get into the hundreds of thousands of users which is a good problem for us to have absolutely right. That's the right time to scale. That is interesting. Though having the bottleneck be the blockchain technology it beings for open source in almost kind of being out of your hands to speed up. Is that scary for you and your team at all? Or how does that work for you guys? It was scary for about a week until I found existing the existing work out essentially so there was a week there of. I don't know if we can do this at all. That was back in your building the MVP so last March or April that required me to really dig through sort of fouls as the Internet being like okay but seriously how did you deal with. And the simplest solution technologically basically. It's the tree data structures where you have part of something onto the blockchain and then you can prove things that branched off of that haven't changed either so it's sort of delegated blockchain technology that solves the problem right now and well. It's usually not considered very elegant. Which is why the mainstream open. Blockchain community is trying to implement something better. It goes work for the sizes that we're looking at which would be in the tens of thousands of documents you know per day or week rather than the millions or the Chilean which is where you get the article from people like you know credit card companies now black hanes not going to work you know. We have billions of transaction. Well that's true blockchain. Technology isn't going to scale to the billions of transactions. That's where they're redesigning it but for the ten thousand we'll be all right for quite right now that makes sense so as you step out on the balcony and you look across block health and everything that you've built. What are you most proud of? I think I'm really proud of the the teams sort of attitude we have. There's three of those difficult to say that you know. How much did that? With my co founders. But I really liked the sort of rapport. We have going with trying to pick up the slack from each other. Non coders being fried with learning how to do some pieces of technical work where they can and you know me trying to make sure I understand the problem on the other side. Ed Really keep that I think is part of our culture going forward the collaborative cultures something. I'm proud of because I think that that had a lot to do with. How sort of flexible we've been able to be and how fast we have been able to progress the software absolutely so. Let's flip the script a little bit. Tell me about a mistake. You made and how you and your team responded to it so honestly the biggest mistake. I've probably made so far as you know. We have worked with several contractors and I think the biggest mistake that is the CPO have made is not interviewing people well. I have made mistakes on both sides. Both on wasting a lot of time looking at candidate who were not able to do the job and then but also ending hiring candidates who weren't actually able to do the job that I asked for them because I use the wrong. I guess scripts nowadays. I probably use the wrong. 'cause you question asking someone. Have you ever used this technology? How did you use it? And then taking people at their word and then maybe not realizing where the true win it so people's understanding With different technologies that I was going to need them to use. Us started or stopped so the problem with that is then you have a contractor. Who can't do the work possibly at all but also you know then your your whole setup is delayed because your product roadmap assumes you are going to be making progress incrementally continuously so. I think by team was pretty nice to me. When they're they would you like? I think we need to pick someone else for that job or I think you need to replace that person in our ad more help so the fulltime members of my team. I think we're pretty patient and gracious when I would say one week. Okay I have a contractor. That means that in three weeks we will have this problem solved and then three weeks later they did not have the problem solved so. I am still struggling with how you really get. Good at that At interviewing process. Yeah so that's definitely an area where you know. I need more sort of experience and working as A. Cto Is really trying to that that the technology people that were trying to get to work with us sure that makes total sense. I think that's an a shared problem across many leaders not to mention. Cto for sure. So what does the future look like for block health for the product and for your team? I mean the future for the product is to really expand the interoperability of it. In the best case scenario we would function like plumbing software right software. Someone might be using one thing. At one end using their companies software at the other end and not realize the blockhouse was the single through the middle there so all of that integration. Though you know your bill that you use the pipelines of those software and That requires a lot of engineers so the future of the product is definitely trying to expand you know interoperability because there's a pitfall for credentialing software for healthcare software in particular in our industry of well we set out to become one thing and we turned into a successful next version of a data silo. So we're just another roadblock in the way that everyone's trying to get the flow through so really the future of our success. I think rise on avoiding that giant trapped because it's such an easy trap to fall into because there's an obvious business model associated with it. Which is well you throw this information in detail decide data silo and then you charge money for it out and just avoiding that and really getting to that next generation software company. Where's selling people's data? It's not your business model is what the future looks like for us. So who influences the way that you work name an architect? Ceo CTO Tech Person. Sure or really anyone that you look up to and what technical person I look up to the most is a former mentor and bosses. Mind so someone I worked with directly at a previous company of his first my then toured leader on when he was promoted and end up ended up asking my boss and I had no idea sort of how good I had it because he was one of my first boxes out of out of school and he was very low key. Say things like. Oh that thing you code it your drawing you design of how it works so that we have it for documentation and use this format and I would and he would give me feedback and I never felt criticized and I was always like oh I should be sharing knowledge. Is this great code that I built then some poets about five years later. I'm mentoring an Internet. A different job getting impatient with intern in my head. And how can I make this person understand what they're building and then it dawned on year? What my mentor had weighed. Like five years prior that he himself did not actually need like me to explain how software work to him like that he actually already understood that but he'd been sort of teaching me how to do all of stat documentation a texture and to think about customer her first and design and he would. He did all of that. You know a lot of patients a lot of his always willing to restate a problem in it. On technical manner for a non technical audience you know he was also technically excellent and that can only imagine himself in other people's shoes and so you know F- I'm trying to figure out how to work. I always want to be like okay. Well what would they do like? You'd be more patients. He would drugged out like you go find the stakeholders just having that sort of like knowledge that nope. There's there's a way to do this. That is like patient because patience is not one of my natural tricks put it lately has been really out both for me. Just have this like muddle to look at both from technical and then also from sort of inner personal perspective. Sure that's awesome so if you could go back to the beginning what would you do differently or consider taking another approach on? I would buy more software. That sounds crazy you know through the CPO AJ. You're supposed to build it but you know it's almost always cheaper to buy the thing you're looking for than it is to build it now. There is a cost to that. You have to integrate it with your existing systems which is often difficult. But I don't think that we spent enough time at the beginning looking at the entire ecosystem of commercial and commercial open source vailable sort of software as services. That are out there. We evaluated them sort of on a constraints basis. Like okay. Well I we pick a database and then we pick a front end and once we've picked those two things it actually eliminates a lot of the other options for for email for message passing or file storage because they'll have to work together and so I think if I'd had a chance or if I ever had the chance to again to go back you know really looking at all of the services that you're going to need and getting a list of every single email provider every single message passing architecture. Every single hugh software every single file storage commercial option available would have been really helpful. Because what you don't realize until you were halfway through the process where you had a Fox. That just said message. Passing architecture. Goes here or third. Party email service goes here. Is that the technical choices you make early on eliminate a bunch of the option. You can't use them anymore. You know they don't work anymore. Answer that is that is something. I would definitely go back a year ago and tell myself you spend more time looking at your sort of folk architectural options and consider buying or of them so you would consider moving essentially buying more software. You're gonNA move faster right. You're not going to have to build everything from scratch. And so you know the engineer and you want to probably. I'm assuming like most of us wants to build everything to our own specification and perfection but the entrepreneur new looks back and sees that we could move a lot faster if we would've bought this this and this is that right. Yeah that's it in a nutshell. Definitely you're getting on a plane and you're sitting next to a brand new. Cto The just built the next big thing and they're pumped about it. There are about to show it to the world. What advice do you give that person? Having gone down this road a bit. It would depend on what they were a CTO in but if they were in the realm of you know call it software as a service in this day and age and they haven't launched it yet. The first thing I would do is ask them if they had Keila on their website just because there are minimum standards that that must be met. But there's more general advice than that. That's literally the first thing that comes to mind. Okay but do you have a feature website? Did you lock down your cloud properly? You know what I definitely learned that over the past year doing this as a CTO was fine. If you yourself are not the networker. 'cause YOU KNOW IN MY COMPANY. I'm not really the network. That's that's my co-founder both. My Co founders are director of operations and credentialing and the. Oh spend a lot more time than I do at conferences on. Podcast even talking to people. If you are not networking person yourself. You have to find them because there are seen in my company that Camille just look like magic like Oh we have venture capitalist pitch to now we have customers to pitch due to demo to now we have pipeline of a hundred people that we're going to that our our CEO is GonNa roll through and you know narrow down to people who would actually buy products. I show up I feel technology. I give them the demo but that is all work and as a engineer who sat in a room with other engineers or in an office with other engineers. I was not properly exposed to that before I worked at a startup where I sat next to the people right before I sat next to the operation people. So if you yourself are not you know that person as a technology specialist or as a first time. Cto You need to find and you either need to become them or you need to hire than Interior Cup because it just have like you hiring people and things happen. I have no idea how we have sixty people that I'm going to pitch to you. But we do. Yes so it can't all be about you or your tax. That's the thing it's not as their thing. That's so hard. I think that most engineers would be willing to check their own ego at the door but they're touches their baby like no. It's not really that your tech either. Like not not about you and it's not about the thing you built. You really need those other people absolutely later and thank you for being on the show today. Thank you for telling the creation story of block health anchor happy and this concludes another chapter of Code Story. Code Story is hosted and produced by Noah lab part season two episodes or co produced and edited by Bradley Dental. Be Sure to subscribe on Apple podcasts. Spotify or the podcasting up your choice support the show unpatriotic dot com slash coach story for just five to ten bucks a month and when you get a chance. Leave us a review both things. Help us out tremendously and thanks again for listening.

Coming up next