 From the CUBE Studios in Palo Alto in Boston, connecting with thought leaders all around the world, this is a CUBE Conversation. Hi, I'm Stu Miniman, your host for Cloud Native Insights. When we launched the series, one of the things we wanted to talk about was that we're not just using cloud as a destination, but really enabling new ways of thinking, being able to use the innovations underneath the cloud, and that if you use services in the cloud that you're not necessarily locked into a solution or can't move forward. And that's why I'm really excited to help. Welcome to the program. I have the co-founders of Vendia. First, we have Dr. Tim Wagner. He is the co-founder and CEO of the company, as well as generally known in the industry as the father of serverless from the AWS Lambda. And his co-founder, Shruti Rao, she is the chief business offer at Vendia. Also came from AWS where she worked on blockchain solutions. Tim and Shruti, thanks so much for joining us. Thanks for having us in here, Stu. Great to join the show. All right, so Shruti, actually, if we could start with you because before we get into Vendia, coming out of Stell, really interesting technology space, you and Tim both learned a lot from working with customers in your previous jobs. Why don't we start from you? Blockchain, of course, had a lot of learnings, a lot of things that people don't understand about what it is and what it isn't. So give us a little bit about what you've learned and help that lead towards what you and Tim and the team are doing with Vendia. Yeah, absolutely. So one most important thing that we've all heard of was this great gravitational pull towards blockchain in 2018 and 2019. Well, I was one of the founders and early adopters of blockchain from Bitcoin and Ethereum space all the way back from 2011 onwards. And at AWS, I started the Amazon managed blockchain and launched quantum ledger database, two services in the blockchain category. What I learned there was no surprise there. There was a gold rush to blockchain from many customers. I personally talked to over 1,092 customers when I ran Amazon managed blockchain for the last two years. And I found that customers were looking at solving this dispersed data problem. Most of my customers had invested in IoT and edge devices and these devices were gathering massive amounts of data. And on the flip side, they also had invested quite a bit of effort in AI and ML and analytics to crunch this data, give them intelligence. But guess what? This data existed in multiple parties, in multiple clouds and multiple technology stacks and they needed a mechanism to get this data from wherever they were into one place so they could use their AI ML analytics investment. And they wanted all of this to be done in real time and they gravitated towards blockchain but blockchain had quite a bit of limitations. It was not scalable. It didn't work with the existing stack that you had forced enterprises to adopt this new technology and entirely new type of infrastructure. It didn't work cross cloud unless you hired expensive consultants or did it yourself and require these specialized developers. For all of these reasons, we've seen many POCs, a majority of POCs just dying on the vine and not ever reaching the production potential. So that is when I realized that what the problem to be solved was not the trust problem. The problem was dispersed data in multiple clouds and multiple stacks problem. Sometimes multiple parties even problem. And that's when Tim and I started talking about how can we bring all of the nascent qualities of Lambda and serverless and use all of the features of blockchain and build something together. And he has an interesting story on his own, right? Yeah, truth to truth, if I could, I'd like to get a little bit of that. So first of all, for our audience, if you're watching this on the main, probably wanna hit pause, go search Tim, go watch a video, read his medium posts about the past, present, and future of serverless. But Tim, I'm excited. You and I have talked in the past but finally getting you on the program. I've looked through my career and my background is infrastructure and the role of infrastructure, we know is always just to support the applications and the data that run business. That's what is important. Even when you talk about cloud, it is the applications, the code and the data that are important. So it's not that, okay, I've got near infinite compute capacity, it's the new things that I can do with it. It's a comment I heard in one of your sessions. You talked about one of the most fascinating things about serverless was just the new creativity that it inspired people to do. And I loved it, it wasn't just unlocking developers to say, okay, I have new ways to write things but even people that weren't traditional coders. I've talked to people in marketing that were like, I can start with this and build something new. So I guess the question I have for you is, we had this idea of platform as a service where even when things like containers launched, it was we were trying to get close to that atomic unit of the application. And often it was talked about, well, do I want it for portability or ease of use? So you've been wrangling and looking at this from a lot of different ways. So is that as a starting point, what did you see in the last few years with Lambda and help connect us up to where Shruti just left off her bit of the story? Absolutely, look the great story of the great success of the cloud is this elimination of undifferentiated heavy lifting from getting rid of having to build out a data center to all the complexity of managing hardware. And that first wave of cloud adoption was just phenomenally successful at that. But as you say, the real thing businesses wrestle with are applications, right? It's ultimately about the business solution, not the hardware and software in which it runs. So the very first time I sat down with Andy Jassy to talk about what eventually became Lambda, one of the things I said was, look, if we want to get 10x the number of people to come and be in the cloud and be successful, it has to be 10 times simpler than it is today. If step one is hire an amazing team of distributed engineers to turn a server into a fault-tolerant, scalable, reliable business solution, that's going to be fundamentally limiting. And we have to find a way to put that in a box to give that capability to people without having them go hire that and build that out in the first place. And so that kind of started this journey for compute, for trying to solve the problem of making compute as easy to use as possible. Take some code, as you said, even if you're not a die-hard programmer or backend engineer, maybe you're just a full stack engineer who loves working on the front end, but the backend isn't your focus, turn that into something that is as scalable, as robust, as secure as somebody who, who has spent their entire career working on that. And that was the promise of serverless, outside of the specifics of anyone cloud. Now, the challenge, of course, when you talk to customers is that you always heard, you always heard these same two considerations. One is, I love the idea of Lambda, but it's AWS. Maybe I have multiple departments or business partners or need to kind of work on multiple clouds. The other challenge is fantastic for compute. What about data? You've kind of left me with, you're giving me sort of half the solution. You've made my compute super easy to use. Can you make my data equally easy to use? And so, obviously, the part of the genesis of, of Vendita is going and tackling those pieces of this, by giving all that, that promise and easy use of serverless, now with a model for replicated state and data, and one that can cross accounts, machines, departments, clouds, companies as easily as it, as easily as it scales on a single cloud today. Okay, so you've covered quite a bit of ground there, Tim. If you could just unpack that a little bit, because you talk about state cutting across environments, what is it that Vendita is bringing? How does that tie into solutions like Lambda, as you mentioned, but other clouds or even potentially on-premises solutions? So what is the IP, the code, the solution that Vendita's offering? Happy to. So let's start with the customer problem here. The thing that every enterprise, every company, frankly, wrestles with is in the modern world, they're producing more data than ever. IoT, digital journeys, mobile, edge devices, more data coming in than ever before. At the same time, more data getting consumed than ever before with deep analytics, supply chain optimization, AIML, to even more consumers of ever more data. The challenge, of course, is that data isn't always inside a company's four walls. In fact, we've heard 80% or more of that data actually lives outside of a company's control. So step one to doing something like AIML isn't even just picking a product or selecting a technology, it's getting all of your data back together again. So that's the problem that we set out to solve with Vendita. And we realized that, and we've got a part of the genesis for the name here, you know, Vendita comes from Vend diagram. So part of that need to bring code and data together across companies, across tech stacks, means the ability to solve some of these longstanding challenges. And we looked at the two sort of big movements out there, two that we've obviously both been involved in. One of the thrillers, which amazing ability to scale, but single account, single cloud, single company. The other one is blockchain and distributed ledgers. Manages to run across parties, across clouds, across tech stacks, but doesn't have a great mechanism for scalability. It's really a single box deployment model. And obviously there are a lot of limitations with that. So our technology and kind of our insight and breakthrough here was bringing those two things together by solving the problems in each of them with the best parts of the other. So re-imagine a blockchain as a cloud native implementation built entirely out of serverless components that have all of the scale, the cost efficiencies, the high utilization, like all the ease of deployment that something like Lambda has today. And at the same time, you know, bring state to serverless, give things like Lambda and the equivalent of other clouds, a simple, easy built-in model so that applications can have multi-cloud, a multi-account state at all times, rather than turning that into a complicated DIY project. So that was kind of our, that was our insight here, you know, and frankly, we're a lot of that, we're a lot of the interesting technology for us is in turning those centralized services, the centralized version of serverless compute or serverless database into a multi-account, multi-cloud experience. And so that's where we've spent a lot of time and energy trying to build something that gives customers a great experience. Yeah, so I've got plenty of background in customers that have the information silos, if you will. So, you know, we know in the unstructured data, you know, so much of it is not searchable, I can't leverage it, you know, shruti, but maybe it might make sense. You know, what is the, you know, would you say some of the top things, some of your early customers are saying, you know, I have this pain point, you know, that's pointing me in your direction. What was leading them to you and you know, how does the solution help them solve that problem? Yeah, absolutely. One of our design partners, our early design partners is this automotive company. They're a premier automotive company. They want, their end goal is to track car parts for warranty recalls issue. So they want to track every single part that goes into a particular car. So there are about 30 to 35,000 parts in each of these cars. And then all the way from manufacturing floor to when the car is sold and when that particular part is replaced eventually toward the end of the life cycle of that part. So for this, they have put together a small test group of their partners, couple of the parts manufacturers, their second tier partners, National Highway Safety Administration is part of this group. Also a couple of dealers and service centers. Now, if you just look at this group of partners, you will see some of these parties have high technology backgrounds just like the auto manufacturers themselves or the part manufacturers. Low modality or low IT competency partners such as the service centers, for them desktop PCs, literally the IT competency. And so it is the service centers. Now, most of majority of these are on multiple clouds. This particular auto customer is on AWS, a manufacturer is on Azure, another one is on GCP. Now they all have to share these large files between each other while making sure that there are some transparency and business rules applicable. For example, two partners who make the same parts or similar parts cannot see each other's data. Most of the participants cannot see the PII data that are not applicable, only the service center can see that. National Highway Safety Administration has read access, not write access. All of that needed to be done and their alternatives before they started using Vendia was either use point-to-point APIs which was very expensive, very cumbersome. It works for a finite small set of parties. It does not scale as and when you add more participants into this particular network. The second option for them was blockchains which they did use. They used Hyperledger Fabric, they used Ethereum Private to see how this works but the scalability with Ethereum Private, it's about 14 to 15 transactions per second. With Hyperledger Fabric, it taps out at 100 or 150 on a good day. Transaction through but it's not just useful. Although all of these are always on systems, they're not serverless. So just provisioning capacity, our customer said it took them two to three weeks per participant. So it's just not a scalable solution. With Vendia, what we delivered to them was this virtual data lake where the sources of this data are on multiple clouds, are on multiple accounts owned by multiple parties but all of that data is shared on a virtual data lake with all of the permissions, with all of the logging, with all of the security, PII and compliance. Now this particular auto manufacturer and the National Highway Safety Administration can run their ML algorithms to gain intelligence off of it and I start to understand patterns of when certain parts go bad or what's the propensity of a certain manufacturing unit producing faulty parts and so on and so forth. This really shows you this concept of unstructured data being shared between parties that are not interconnected with each other while when there are data silos. But I love to follow this up with another example of the democratization. Democratization is very important to Vendia. When Tim launched Lambda and founded the AWS serverless movement as a whole and at AWS, one thing, very important thing happened. It lowered the barrier to entry for a new wave of businesses that could just experiment, try out new things. If it failed, they scrap it. If it worked, they could scale it out. That was possible because of the entry point, because of the paper use and the architecture itself. And our vision and mission for Vendia is that Vendia fuels the next generation of multi-party connected distributed applications. Our second design partner is actually a nonprofit that is in animal welfare industry. Their mission is to maintain a no kill for dogs and cats in the United States. And the number one reason for overpopulations of dogs and cats in the shelters is dogs lost. Dogs and cats lost during natural disasters like the hurricane season. And when that happens, and when let's say your dogs get lost and you wanna find the dog, the ID or the chip reading is not reliable. You wanna search this through pictures. But we also know that if you look at a picture of a dog, four people can come up with four different breed names. So, and this particular nonprofit has 2,500 plus partners across the US and they're all low to no IT modalities. Some of them have higher IT competency and a huge turnover because of volunteer employees. So what we did for them was came up with a mechanism where they could connect with all 2,500 of these participants very easily in a very cost effective way and get all of the pictures of all of the dogs and all these repositories into one data lake so they can run some kind of a dog facial recognition algorithm on it and identify where my lost dog is in minutes as opposed to days it used to take before. So you see a very large customer with very sophisticated IT competency use this also a nonprofit being able to use this and they were both able to get to this outcome in days not months or years as a critical blockchain but just under a few days. So we're very excited about that. Thank you so much for the examples. All right, Tim, before we get to the end I'm wondering if you could take us under the hood a little bit here. My understanding the solution that you talk about if it's universal apps or what you call Unis, I believe. So if I saw that right, give me a little bit of a comparison contrast if you will. Obviously there's been a lot of interest in what Kubernetes has been doing. We've been watching closely, there's connections between what Kubernetes is doing and serverless with the K native project. When I saw the first video talking about Vendia you said we're serverless and we're even containerless underneath. So help us understand because at a super high level some of the multi-cloud and making things very flexible sound very similar. So how has Vendia different and what do you feel your architecture helps solve this really challenging problem? Sure, awesome, look one of the tenets that we had here was that things have to be as easy as possible for customers and if you think about the way somebody walks up today to an existing database system, they say look I've got a schema, I know the shape of my data and a few minutes later I can get a production database. Now it's single user, single cloud, single consumer there but it's a very fast, simple process that doesn't require having code, hiring a team, et cetera. And we want it Vendia to work the same way. Somebody can walk up with a JSON schema handed to us five minutes later they have a database only now it's a multi-party database that's decentralized, so runs across multiple platforms, multiple clouds, multiple technology stacks instead of being single user. So that's going to go one is make that as easy to use as possible. The other key tenant though is we don't want to be the least common denominator of the cloud. One of the challenges with saying everyone's going to deploy their own servers, they're going to run all their own software, they're going to build, they're all going to go deploy a Kubernetes cluster. One of the challenges with that is that Ashruti was saying first anyone for whom that's a challenge, if you don't have a whole IT department wrapped around you that's a difficult proposition to get started on no matter how amazing that technology might be. The other challenge with it though is that it locks you out, it's sort of the inverse of the lock-in process, right, it's the lock-out process. It locks you out of some of the best and brightest things that the public cloud providers have come up with. And we wanted to empower customers to pick the best degree, maybe they want to go use IBM Watson, maybe they want to use a database on Google and at the same time they want to ingest IoT on AWS and they want to all to work together, they want all of that to be seamless, not something where they have to recreate an experience over and over and over again on three different clouds. So that was our goal here in producing this. What we designed as an architecture was decentralized data storage at the core of it. So think about all the precepts you hear with blockchain, they're all there, they all just look different. So we use a NoSQL database to store data so that we could scale that easily. We still have a consensus algorithm only now it's a high speed serverless and cloud function based mechanism. Instead of smart contracts, you write things in a cloud function like Lambda instead. So no more learning solidity, now you can use any language you want. So we changed how we think about that architecture but many of those ideas that got people about it really excited about blockchain and its capabilities and the vision for the future are still alive and well. They've just been implemented in a way that's far more practical and effective for the enterprise. All right, so what environments can I use today for your solution? Shruti talked about customers, spanning across some of the clouds. So what's available kind of today? What's on the roadmap in the future? Will this include beyond this, maybe the top five or six hyperscalers, can I do? Does it just require serverless underneath? So will things that are in a customer's own data center eventually support that? Absolutely, so what we're doing right now is having people sign up for our preview release. So in the next few weeks, we're gonna start turning that on for early access to developers. That'll be the early access program, will be multi account focused on AWS. And then end of summer, we'll be doing our GA release which will be multi cloud. So we'll actually be able to operate across multiple clouds, multiple cloud services on different platforms. But even from day one, we'll have API support in there. So if you've got a service, could even be running on a mainframe, could be on-prem. If it's API based, you can still interact with Vendia and still get the benefits of the system. So developers, please start signing up. You can go find more information on vendia.net and we're really looking forward to getting some of that early feedback and hear more from the people that were the most excited to have start building these projects. Excellent, well, great call to action to get the developers and the users in there. Shruti, if you could just give us the last bit, how the thing that's been fascinating, Tim, when I look at the serverless movement, I've talked to some amazing companies that were two or three people and out of their basement and they created a business and then they're like, oh my gosh, I got VC funding and it's usually sub 10 million dollars. So I look at your team, I had heard, Tim, you're the primary coder on the team and when it comes to the seed funding, it's compared to many startups, it's a small number. So Shruti, give us a little bit, if you could, the speeds and feeds of the company, your funding and any places that you're hiring. Yeah, we are definitely hiring. Let's me start from there. We're hiring for developers and we are also hiring for Solution Architect. So please go to vendia.net and we have all the roles listed there. We would love to hear from you. The second one, funding. Yes, Tim is our main developer and solutions architect here and look, the serverless movement really helped quite a few companies, including us to build this, bring this to market in record speeds and we're very thankful that Tim and AWS started taking the stands back in 2014, 2013 to bring this to market and democratize this. I think when we brought this new concept to our investors and they saw what this could be. It's not an easy concept to understand in the first wave, but when you understand the problem space, you see that the opportunity is pretty endless and I'll say this for our investors on behalf of our investors that they saw a real founder market fit between us. We're literally the two people who have launched and been brand businesses for both serverless and blockchain at scale. So that's what they thought was very attractive to them. And then look, it's Tim and I and we're looking to hire eight to 10 folks and I think we have gotten to a space where we're making a meaningful difference to the world and we would love for more people to join us, join this movement and democratize this dispersed data problem and solve for this and help us create more meanings to the data that our customers and our companies worldwide are creating. We're very excited and we're very thankful for all of our investors to be deeply committed to us and having conviction on us. Well, Shruti and Tim, first of all, congratulations. Thank you, thank you. We've done absolutely looking forward to watching the progress going forward. Thanks so much for joining us. Thank you, sir. Thank you very much. All right, and definitely tune in to our regular conversations on cloud native insights. I'm your host Stu Miniman and looking forward to hearing more about your cloud native insights.