Michael Ames [00:00:19] Hey, everybody, happy to welcome you today to this episode of SADA Cloud N Clear, where we bring in industry thinkers, great leaders and great engineers to talk to us about how they’re changing the world with Google Cloud Platform. We’ve got some great guests today and really excited for you to get a chance to hear from them, get to know them just a little bit. Co-interviewing with me today, I want to introduce Simon Margolis, VP of Cloud Engineering for SADA, responsible for our amazing team of technical engineers and consultants. And also really excited today to have a couple of folks from Vida Health who are in, we’re going to let them tell you about their business, in the digital health and in the business of helping people improve their lives and helping companies to manage health care costs with the technology that they’re building on Google Cloud Platform. So I’m going to give them each a chance to introduce themselves and to tell a little bit about their background, what brought them to Vida, and then we’re going to dig in with questions about their business, their technology and where they’re going next. So let’s start with Amol. Amol Kher thank you for being here. Tell us a little bit about yourself.
Amol Kher [00:01:35] Thanks, Michael, and thanks, Simon. I’ve been at Vida for about a year now, and one of the main reasons why I joined Vida is that we are a virtual chronic care platform, digital health platform for members to come in and get the help from a provider to treat any of the chronic conditions they have. So things like mental health, diabetes, hypertension, that sort of stuff. And so one of the biggest sort of motivators for me to join Vida compared to some of the other folks was that Vida takes a very unique approach it being a poly-chronic, we call it poly-chronic, which is we treat multiple conditions at the same time. It’s a differentiator in the market compared to say some of the vertical players that only focus on diabetes or only focus on hypertension or stress and stuff like that. It’s a core philosophy that Vida believes in, or the founders of Vida believed in, is that members are poly-chronic, which is like when you suffer from one condition, you typically suffer from multiple conditions. And so, just telling you like, hey take some diabetes medicine, doesn’t necessarily address the whole you. And so that’s what drew me to this platform. It came from my background, which is that I had done a startup about eight or nine years ago, which is in the fitness vertical. And I often saw that when people tried to get advice, they would also get into nutrition as an example. So it’s not just about what exercise you do, but what do you eat and that sort of stuff, because oftentimes they’re trying to do weight loss as an example. So I knew that just because the person came to you with a problem doesn’t mean that’s the only problem that they’re dealing with. And having a platform that can actually help them look at the entire spectrum is actually gives better outcomes, too, for the members. So that’s what drew me to Vida and that’s why I’m here. And then, of course, I get to work with brilliant people like Ty and I will let him explain why he joined as well.
Michael Ames [00:03:27] Yeah, go ahead. Tell us a little bit about your background and what brought you to Vida.
Tai Yu [00:03:31] My background is actually from a networking background at Cisco for a decade and I’ve transitioned into a full stack Python developer specializing in platform infrastructure. And at Vida we are the core competence of Vida is to bring members and providers together. Connecting them through chat, video, there’s a lot of communication aspects to this to get the best results for our members through remote care. And the passion I have is to really build a efficient, scalable and performant platform to make that vision a reality for our members, providers, investors and our coworkers as well. This is more than just work. I think it benefits a lot of people, their health and their futures. So I think it’s more of a feel good type of industry to be working with these great people.
Michael Ames [00:04:38] Great. Thank you, Ty. What draws a lot of us into the health care industry is that idea that we’re doing work that’s ultimately going to help improve people’s lives, help them lead better, fuller, richer lives. So we appreciate the work that you’re doing. Simon, why don’t you jump in?
Simon Margolis [00:04:53] Yeah, absolutely. So thank you for the introductions. I think this is going to be an exciting conversation. Given my role, I want to jump right into the nitty gritty technical stuff. But obviously there’s a reason you ended up selecting cloud computing in general and reasons that brought you to Google and to SADA to pursue your cloud journey. One of the two of you or both of you tell us a little bit about what did that journey look like, what drove you towards Cloud, Google Cloud, SADA and all the successes that you’ve had so far?
Amol Kher [00:05:24] I’ll take a stab at it. Healthcare as an industry is pretty enormous and already fairly complicated. And for a startup to be playing in this space, going against some entrenched players and entrenched processes, one of the problems we don’t want to have on our plate when you’re building a team in a company is how do you manage your servers? And how do you manage all the hardware, upgrades, security, compliance and that sort of stuff? And so for us, like cloud is a no brainer. We’ve always been in the cloud. And you could say we are cloud first in some ways. So to me, it makes business sense as well to not try to – we want to innovate more on the delivery models and how we, as Ty was pointing out, how we connect members with providers and not have to worry about the nitty gritty of the infrastructure and the machines and that sort of stuff.
Michael Ames [00:06:12] As you talk about launching a startup and not wanting to have to be concerned with things like servers and data centers. It shows how much the industry has changed. That, I was with a startup 20 years ago and part of what we did to try to lure investors is we would bring them into our office space and we take them back to the server room because they could see all of those shiny boxes and wires and the flashing lights. And that was how we convinced them that we were a real company who actually knew what we were doing. How much better is it today that they assume that you’ve got that part figured out? If you’re not launching on the cloud, you’re probably wasting their time. What they actually want to know about is how are you going to use that technology to start to deliver value, deliver it to your customers? So it’s just a fascinating way that the startup industry has changed. The thing that was the selling point before now is something you need to show that you’ve actually moved beyond that you’re ready to start delivering value with cloud technology.
Simon Margolis [00:07:08] And obviously, there’s a lot here that we at SADA are obviously a bit biased towards Google Cloud, we think it does a lot of things really well, but maybe what were the top one or two reasons that you selected Google to ultimately develop the remainder of your platform on?
Tai Yu [00:07:22] We were actually already on Google. We have a big data footprint, BigQuery we were using for our analytics and machine learning before. So moving from Aptible to Google was kind of a no brainer because of that inertia there. And we feel Google provides us a great, scalable platform where we can deliver features and functionality very quickly in a single dashboard platform. So with our business, we need to grow and scale fast. And the auto scaling capabilities of Google to be able to scale to millions of users without actually changing much of our code and infrastructure gives us an edge compared to other entrenched server farm companies. So I feel that Google was one of the two candidates for migration and we felt we made a good choice in going with Google.
Amol Kher [00:08:30] Yeah, I think one other part besides the analytics and BigQuery stuff is that you have a large provider organization. We also had a Google suite customer, right? So all our Gmail, email, calendar, docs, all of that stuff also is on Google. And actually, one of the things we’ve seen now is that there’s a surprisingly tight integration between all of these systems as well that actually helps us build better processes going forward, which we could have done them with other external tools, but then that would have taken a longer time. And for us, again, as a smaller company, we want to be more efficient. And having all of that data in one place is very powerful.
Simon Margolis [00:09:11] Yeah, very good. Yeah, this is something we hear from a lot of our customers. There was already a data gravity that was pulling towards using GCP. Like Ty had mentioned, using BigQuery is obviously a beautiful Googles solution that does a lot to deliver value. I do want to ask you a little bit more about what you’re doing and how you’re using these tools today to actually bring value.
Michael Ames [00:09:30] Let’s move back up a little bit and just talk a little bit more about what it is that Vida Health does. There’s a consumer facing app that folks can go and download from the App Store. I did it over the weekend, spent some time with it, had a great experience. But consumers aren’t necessarily your direct purchasing customer. Tell us a little bit more about Vida, how you operate as a business and what this product is.
Amol Kher [00:09:54] So I think you saw the consumer facing part of it, which is the members downloading the app. The other part is the providers. There is a web platform for that so the providers have very similar to what you would see if you were walking into an actual in-person clinic where, you know, the provider, the doctor is taking notes and scheduling meetings and stuff like that. So they have a portal, we have a web portal to do all of that stuff. And then, of course, there’s a piece that is more of the configuration and enterprise configuration side of stuff. So when we onboard a new employer or a new peer into our system, we kind of configure information about them and their specific sort of rules that they want set up. Like, as an example, could be things like what is the pricing for some of these things or what other different programs that can be offered to their employees? Because company A might say, we only want to sign up Vida for mental health and IBDs. And another one might say, we want to sign up for mental health and weight loss and stuff like that. And so we have a system that sort of can pick and choose those programs and say, OK, any employee that enrolls from this company. So there’s essentially three different sort of big systems in the team and they all share. And that’s why the data gravity on Google is so important, because at the end of the day, all of that data interacts with two different parties, right? Sometimes a member, for instance, members can sometimes add devices that they are using to track steps as an example. So if they’re in a weight loss program, the coach might say, hey it’s important for you to track your food or it’s important to track your exercise activity. They might use a device like Fitbit. We connect that and then it goes into a database, but then the providers have access to that information. It’s these kind of things, so it’s three different sort of systems that we have.
Michael Ames [00:11:49] Yeah, here’s what I think is exciting about that is you guys are part of this growing wave of digital health transformations where you’re putting tools into the hands of consumers, of basically patients, empowering them to do more things, to take care of their own health. But there’s this whole other side of it, which is that the doctors are actually involved and the payers are involved. You’re involving the employers. You’re involving the health plans. You’re not just creating a cool health app, which there are lots of. And I use them, right? And it’s fun to go download them and play with them. But in the end, you’ve seen a need to connect a lot of different people in organizations together, the patients, the payers, providers, the health plans. And any one of those people doesn’t see all of this. All of that’s got to be supported by a backend that can integrate all of that data, help you to do things useful to it, and help you with one engineering team to support at least three different user interface applications that three different customer bases are logging into to make all of this work together. That’s a really cool, I think, kind of industry solution that you have. And I am interested, Simon, if you want to dig in a little bit more as to some of how you have put this all together in the back end.
Amol Kher [00:13:07] Yeah, I think one thing I want to say, like that’s also a unique thing about Vida, is that we do emphasize the human in the loop aspect of it, because there are a lot of companies that are either just doing more machine learning-based recommendations or they’ll just give you helpful tips on how to improve yourselves, or just content-based programs where it’s like, hey I want to lose weight. Here’s a 12-week content program for you, which is first week you watch this video, second week, you watch this video. It’s all asynchronous, right? Someone’s recording those videos and just sending it to you. In our case there’s actually providers that you’re meeting with and they are interpreting your results for you and telling you last week was a bad week, last week was a good week, and this is what you should do next week. I think it’s important to remember that there’s a lot of synchronous communication going on as well on the platform.
Tai Yu [00:13:53] I’d like to add that we’re bringing the latest technology to bear to this health care with digital health. And with these capabilities, members can send a chat to our provider 24/7, log food intake, log exercise through their Apple Watch and digital glucometers and other devices. And it’s an order of magnitude different compared to traditional care where they only are scheduled in the clinics or in the hospitals. So you are able to meet remotely at any place with either just a phone or video. And so for all the content that members are receiving, they are gotten offline and they can be reviewed in their leisure. So they’re learning, they’re practicing and we’re monitoring their improvements and gauging. So the outcomes that can be had with all these technologies and interactions is much, much better than the traditional way of providing health care.
Simon Margolis [00:15:04] Yeah, completely. And I think that’s an obvious differentiator that your platform provides, which I think is frankly pretty cool. I do want to dive into something you just touched on, though, which is around providing data to and from providers, the end users, and so on and so forth. You had also mentioned the beginning here that you were using already BigQuery on Google Cloud prior to moving the rest of the platform over. Tell me a little bit about what you’re using in terms of Google’s data features today to provide this experience to the various parties that you serve.
Tai Yu [00:15:35] So we use Cloud SQL for Postgres database, so MySQL as well. We have Memorystore that replaces Redis, and Elasticsearch for our searching engine. It’s a third party tool that we’ve integrated using GCP marketplace. All these managed services provides us the ability to deploy and scale automatically via Google. And we don’t really need a dedicated DevOps staff to do this. And it provides us the capability to upgrade and upscale our technology and capabilities naturally as we grow in business and our technological prowess. So we also integrate with Twilio for our chat and video conference capabilities. But within GCP, it allows us that auto scale capability that we didn’t have before in Aptible. This allows us to efficiently scale our pods and be able to decrease our container’s usage at nighttime when we have very low usage and all the monitoring is done via Datadog. And we’re able to monitor the system 24/7 with alerting and metrics. So it really gives us a very, very well managed and centralized platform for us to grow and efficiently deploy services for any new features that we can think of. So it gives us the ability to grow and improve and add features as we see fit for the industry.
Amol Kher [00:17:30] So I was just going to say on the BigQuery side, one of the other big investments we’ve made is also investing in a BI tool like Looker that then can be centrally used across the organization with like our different analysts in the team is one of the things that Vida is I think unique. Again, on the data side is that we have information about claims from medical claims from people. We have information about eligibility data, whether someone is eligible in this particular insurance or not. We then have this app health data. We have engagement data, which is how are they performing? And then finally, we have this outcomes data as well, which is our own how to improve the thing? So we like to think of it as like we have the 360 view of all of the person and having all of this data housed in one single place as a BigQuery allows us to do, I think, more end-to-end analysis of how our patient population is moving and what trends are we seeing and that sort of stuff. So I think that’s another big factor in us choosing the cloud and Google in particular, because that’s where we were already housing.
Simon Margolis [00:18:39] Yeah, that’s awesome. And that’s what I was wondering, and thank you for answering that, is we do have plenty of customers that take advantage of all of the different types of data sharing that Google can allow for, right? Helping reduce data silos and things like that. It sounds like you’re getting benefit from that. But it’s also interesting to see how you’re using these tools to do analytics and to gain insights with tools like Looker to better understand what kind of data you are sitting on, what you’re collecting, and what you can do to further the business as a result of that. It’s really cool to hear. Really interesting how you’re using that.
Michael Ames [00:19:09] Ty, I’d like to ask a follow up question you mentioned moving from Aptible, Aptible is a great platform and one of its selling points is a certain amount of compliance out of the box, right? It is designed to accelerate you toward things like HIPAA compliance. So as you moved from Aptible to GCP, there was some work that you needed to do in order to backfill those compliance capabilities or to at least reengineer those compliance capabilities on GCP. Is there, can you talk to us a little bit about that experience? How did you find building for HIPAA compliance on Google Cloud Platform?
Tai Yu [00:19:45] GCP is already HIPAA compliant today. There are a list of their functionality and infrastructure that is HIPAA compliant and we have that list. BigQuery is one of them, Cloud DNS, Cloud SQL, Kubernetes, and Secret Manager. All these features and functionality that we depend on have already been audited and been shown to be HIPAA compliant. This wasn’t the case when we started to use Aptible, but now it is. And in terms of moving the data from Aptible to GCP, it was pretty much a straight forward lift and shift and be cognizant of the security of the data during the transfer. But it wasn’t that much of a difference in terms of worrying about the GCP HIPAA compliance. We felt comfortable with GCP in that regard. And we created a bastion host, we made sure that the link between the two Postgres instances, and we created a read replica to make the transfer faster and secure. We were able to do this in one night, one Saturday night. And this was all due to SADA’s help and their consultants. And they were able to do this and be able to be agile and flexible in their timing and abilities to do this for us.
Michael Ames [00:21:29] Ty I appreciate that. And so, you know, a couple of things that were good to hear, just sort of reflecting back again, looking back five, six years ago, we were having conversations in the industry like, can we be compliant on the cloud? And it felt like we got past that to the answer is yes. And then a few years ago, the question became, OK, we can be compliant, but how hard is it going to be? And it’s exciting that we hear more and more stories like yours that reflect that the out of the box compliance that we’re getting from Google Cloud is actually already an accelerator. You had additional work to do. You guys followed some great practices with bastion hosts and other kinds of things. But the lift to get there wasn’t onerous. And I think that’s important for people to understand. I also love the startup lifestyle that it only took one Saturday night to get it done. Ty your last comments there about SADA and consulting are a good segway, we would love to hear from you and from Amol why it was ultimately that you chose SADA. There are a lot of different Google Cloud partners out there. We know why we like us, but we’re curious about that decision and how well that is working out.
Amol Kher [00:22:37] So I met with my Google counterpart on our account manager, Google, Will. My previous company we were on AWS, and AWS does preferred services and you can get a account manager that can help you do some of this, like long term contract negotiations. So I came into this from that angle, not fully knowing about the way Google works with partners like SADA. So as I worked with Will, he kind of explained like, you know, at the startup level, that’s how Google prefers to operate. And he introduced me to three or four different vendors in the space. And one of the things that I asked him was like, I want to see three different vendors who are maybe different sized, but also people who have like reach or understanding of compliance and migrations, as an example. And then one of the partners he recommended was SADA, obviously. So I had a great initial meeting. Of course, we do our own back channel references and there were great references from other people who had worked with SADA. And ultimately, once I worked with the account manager on your side, it was Anthony, and it seemed like a good fit from a point of view of the skill set that you guys can bring in, which is all around migrations, but also ongoing sort of support that you offer around giving us a preferred line into like Google. As a small company, it’s really hard to get Google’s attention. But if you’re pooling in with someone like SADA who works with multiple different customers, then you get more attention. Try to solve a problem, and it happens. So taking into account all those factors, which is like SADA has a lot of existing customers, you guys have done migrations, you’re looking for a partner who can help us with the migration as well, working with someone who can help build a long term contract, provide the ongoing customer support. I think you guys checked all the boxes and of course, the references were good as well. So that’s ultimately how I ended up making the decision. And then Ty can put his points on like any things he sort of evaluated from a technical side of stuff.
Tai Yu [00:24:40] For me, SADA has really brought the best industry practices and the experiences that they’ve gained from working with many big companies over the years. And they recommended Helm for our deployments and Terraform for deploying the infrastructure. And those two really key advice, which we followed through, really gave us a good footing and solid foundation to build on. And with their professionalism, the timeliness and predictable results that pretty much guaranteed that the migration to be done in four weeks. We feel that accountability was worth more than we actually paid. So that predictable, HIPAA compliant, guaranteed migration is what we really value in a business like ours. It’s great.
Amol Kher [00:25:34] One of the references that you might find amusing is that one of the executives that I talked to who had used SADA before said that oftentimes the people at SADA that he worked with are better than his internal folks. And so he ended up working at SADA for a number of years because of that. And that sort of gave me some measure of confidence, because oftentimes when you’re working with a third party, you’re not sure what the level of talent is that you’re bringing in. And we have a small team. It’s fairly talented, not deeply invested in like the DevOps side of stuff. So I needed to bring someone who has the gravitas and has the experience have done this multiple times over. Just thought that was an interesting thing that sort of convinced me to say, let’s go with this partner.
Simon Margolis [00:26:17] I’m glad to hear it. Obviously, this is one of the things I care the most about is making sure that we do provide the folks from the engineering side that have seen this stuff before and have been through it before. So it sounds like you had that experience, which is great to hear. I’d ask you for more detail, but the fact that you were able to do sort of a overnight go live and have it be successful, I think speaks to the entirety of the project. So that’s really awesome to hear. I love seeing results like that and I’m really happy that you were able to do that. Of course, though, a migration to Google Cloud Platform is a means to an end, right? It’s not the end of the project. In fact, in many cases, that’s the beginning of your cloud journey. So I’m interested to ask actually both of you, maybe Ty we can start with, a little bit about what the future holds. Both in terms of the immediate near term, as well as maybe some loftier aspirations as to what you want to do now that you’re in such a scalable and performant environment.
Tai Yu [00:27:11] Yeah, at the beginning of the engagement with SADA, we really considered Autopilot GKE. And Autopilot GKE is a brand new service by Google where you’re not needing to manage your own nodes and you’re paying for the CPU and memory requested. And it really helps us manage the GKE a lot better, easier, it’s more maintainable and its default secure. So that’s our next steps. Unfortunately, we had a hiccup with Datadog supporting Autopilot in the first versions, but the next couple of months they’ll release a version that has fully support. And after that, we are probably going to use the latest and greatest of ElasticSearch with Kibana to up level our search capability. And investigate GCP’s API Gateway to really centralize the authentication, the firewall, and load balancing and helps us get our foot into the server lists backend. We want to explore that to be able to have it be even more scalable, more efficient use of our back-end services. But this is long term horizon. We haven’t really nailed down anything yet, but the Autopilot GKE looks very promising as our near-term future, where we want to end up.
Simon Margolis [00:28:40] Yeah. Of course, yeah. And it makes sense, right? It sounds like you’re doubling down on a lot of the things that brought you to GCP to begin with. The automation, the – I don’t want to say set it and forget it, but some of the things that allow you to do all that scale with less human intervention. It sounds like you’re going to do more of that. For what it’s worth, GKE Autopilot is really cool and I think that’s going to be a really fun one for you to play around with. But Amol, I wanted to hear maybe the same from you as well as those longer term sort of strategic initiatives that you hope to accomplish.
Amol Kher [00:29:07] Ty touched on a lot of the technical architectural sort of points. I would say, from a product point of view or feature set point of view, one of the things that’s exciting about using something like Google now is Google is a massive company and they’ve invested a lot in their more of their vertical health care APIs as well. I think one of the reasons why Vida went into Google in the first place along with Aptible was the, I think Google had a slight edge in the machine learning AI space. Vida, as a company, we believe that to really reach scale we have to get, you know, great outcomes very efficiently at low cost. Typically, when you have a human in the loop involved, costs can be high. And what we want to do is we want to build machine learning models that can augment our providers to make them more efficient as they’re taking notes or, you know, doing things like that. We plan to use Google Cloud’s growing capabilities in the healthcare API space to build stuff now on top of that. So it’s like we’re getting more cloud native, using more of what we have, as you said, double down. We put our chips all in and so we going to keep calling bets on that one, essentially.
Simon Margolis [00:30:20] Yeah, very cool. And I think, again, those are some of the coolest things that our customers are able to do, leveraging the technology that Google has, the ML, the A.I. is, I think, a couple of steps ahead of their competition, you can see that. The consumer Google products, right? Google.com, Gmail, YouTube, so on and so forth. You can see that. And I think that, and maybe I’m a little biased because my audience here. But I think that some of the things that you can do in terms of health care with those technologies, I think is some of the most exciting to see what the future is going to hold. I know I’ll definitely be closely following this adoption curve for Vida and see what you guys are able to do with it. I think it could be really great.
Michael Ames [00:30:53] I just want to thank you for coming and spending time with us today talking about the Vida story, talking about your stories, talking about Google, how you got to where you’re at, where you’re going next. Encourage anyone who’s listening to head over to Vida.com, V-I-D-A, and go learn about these products. If you’re running a business and you are interested in reducing costs and improving health for your customers, you can engage with them at the business level. If you’re just an individual though, who is interested in the polychronic, multiple, multi-symptomatic sort of approach to disease management, go download their app and play with it on the weekend. I think there’s a seven day free trial and some fun things that you can do there. And I got to say Amol and Ty, we, speaking for SADA, are so excited to be aligned with you on your journey here. The technologies that you’re intending to roll out, the health care value that you’re trying to put out there in the world, we’re there for you side by side, excited to see you go to great places and to do whatever we can do to help you get there. So thanks, everyone, for joining and see you next Cloud N Clear.
Amol Kher [00:32:05] Thank you so much for your time.
Tai Yu [00:32:06] Thank you.
Narrator [00:32:09] Thank you for listening to Cloud N Clear. Check the show notes for links to this week’s topics, and don’t forget to connect with us on Twitter @cloudnclear and our website, SADA.com. Be sure to rate and review the show on your favorite podcast app.
