 Hello, and welcome to this special CUBE Conversation. I'm John Furrier, host of theCUBE here in our Palo Alto studios. I'm here talking about business transformation. We had a great story here between persistent and blast motion partnering with Genevieve AI as the time has shifted where the value of the data is super important and we're going to see it in action. Dr. Baskar Bose, Vice President of Advanced Engineering at Blast Motion Inc. And Dr. Pandarang Kamat, who's the CTO of persistence, talking about Genevieve AI and blast motion and persistent working together. Doctors, gentlemen, thanks for coming on this CUBE Conversation. Thank you for having us here. Yeah, thank you. Love to get the PhDs up, put the doctor on there. It's just great accomplishment. But you know, we're in an era now where democratization of data is coming down to all companies. You guys have a special relationship and the timing of it is super important. Blast motion, Baskar, talk about your company before we get into the persistence side of it. You guys have a unique opportunity with your data at this point in time. It's not just you thought about it this morning. It's been around for a while. Take a minute to introduce Blast Motion. Yeah, so Blast Motion is one of the technology leaders in motion capture technology. We put sensors on sports equipment and we collect data and we push that data through our mobile app up to our cloud and we provide advanced insights based on that data. The goal has always to focus on the athlete and make them the best versions of themselves. So that unique data is primarily for the particular athlete themselves and we call that personalized actionable insights. So you guys are taking motions of people's golf swings, baseball sports, mainly, right, mostly sports? Yeah, I mean, the way we've developed the system, you know, the sensor technology is, you know, developed for high volume manufacturing. We have proprietary calibration methodology so that each sensor coming off the factory line calculates the exact same data. And this technology goes across, you know, all levels of play. So it's the same sensor between what the pros use versus what the youth player uses. And so the idea is to collect data on every swing and then be able to provide insights on that. So for context, how many years you've been doing this? How much data have you gathered? Take us through some of the timetable here and what was your storage solution? Because now the time has come, the data values high right now in AI, that's where the gold is as they say in this market is quality data. How long you've been doing it and how much data have you acquired? Yeah, so we've collected over a decade of data across all levels of play. And, you know, from the very beginning, you know, what we did was we designed the sensor, the mobile apps and the clouds to be domain agnostic. So it was primarily capturing motion data. And we have over 300 million swings and, you know, we're on track to collect another 100 million swings this year. And every swing represents a player basically taking a swing or a golf stroke or, you know, jumping. All that kind of data is being collected in the cloud. And so one of the visions that we had very early on that this data is gonna be valuable and that we would mine that data for insights. And so we've been waiting for AI to come along because it's a lot of data and it is complicated. And so we need a system that can identify patterns in the data, develop insights from that data. Now, part of the groundwork that we've been doing for the last, you know, several years was, you know, we have high quality data scientists on staff. And so they've been analyzing the data manually and we're taking that knowledge and transferring that into a generative AI system. Hey, Daron, talk about the persistence side of it. You're the CTO. I mean, you got to be looking at this saying, wow, this is the timing's perfect. Foundational models are hot. You got a great relationship with AWS. They got infrastructure and the technology and data. You guys have a deep experience in generative AI and machine learning. Talk about how this came together on your side because I can imagine when you saw this, the genera AI trend was perfect here for what they got. The foundational models are booming. Really good position to develop this out further. Tell your piece of it. Sure. So yeah, like you said, I represent persistent. We are the fastest growing digital engineering company in our segment, headquartered in India and US. And our pedigree always has been deep tech and product engineering. So that's one of the unique differentiators that we bring to our partners and customers. And we have been working with AI for a very long time now. That's one of our longest running practices. So about nine months ago, we began our generative AI journey, pivoting very targeted way and focused way. What we realized early on was, look, while the hype at the time was all around consumer grade applications and what people were using, for our customers and our business, what mattered was bringing generative AI to life in an enterprise grade, enterprise safe, and enterprise scale way. That meant working with the right security models, right data models, working with our hyperscaler partners like AWS and bringing those solutions to life for our customers. So when the blast motion opportunity came along and we had been working with them in helping modernize some of their data to the cloud, this was a perfect thing where we could take the very unique and very extensive data set that they have, which is a very unique mode for them. And taking that, coupling that with generative AI and figuring out how we can make a meaningful difference to the lives of the players and at all levels who are using their technology. And we are very excited about this journey. Baskar, talk about the blast IQ. This is the next phase of your blast motion journey. Give a quick description of what that is. This is where you're starting to see some actual insights. Can you give a quick update on the blast IQ? Yeah, so we've been developing products for many years and working with our customers. And everything from youth athletes, to parents, to coaches, the biggest questions are, what do the metrics mean? What's good and how do I get better? So those are the top three questions. And as we've worked through the player journey, one of the things that, we're sort of transforming all our applications over to a journey-based application so that we can answer those questions. We wanna get away from being a data-centric company to being more information-centric and then guiding the user. So that's why this partnership with persistence is really important because generative AI is a key component to communicating our technology to our various personas that we're working with. Pennar, talk about the getting started with the blast motion, January. How did you guys put this together? Obviously the swing data repository's there. Really good position to kind of leverage the data and kind of change its trajectory. How has persistence experience, what did you guys bring to the table here? I mean, there's different stages of generative AI adoption here. How did you guys help them get started? Sure, it all starts with this, the blast motion sensor and the data it collects and the variety and the accuracy with which it does that. But the generative AI story doesn't begin and end with generative AI. It actually begins with a lot of the data wrangling that one does with traditional data technologies and AI technologies. So one of the first things is around figuring out the baseline, analyzing all these 300 million swings and figuring out baselines for various parameters that's using your traditional tools around that. The second part then is the first MVP that we wanted to get started with is how to help players become the best versions of themselves. So for that, you need to do time series analysis of that player's swings. How they are doing themselves over time, how they are swing acceleration angles is changing over time, are they improving or not? The second thing is then to compare not only against the baselines but compare against their own goals, figuring out how we can help them get there and figuring out what are the right drills from the extensive data that blast motion has. How can we recommend those in the right way? So all of this require multiple techniques whether it's clustering, whether it is time series analysis, whether it is data cleaning and then you bring the generative AI angle to add the very human-esque touch to it. And I say human-esque because it can't replace a human. But how can our mission together was really how do you put an expert biomechanics coach right in the pocket of every player with this? And how can we not only present insights and data as graphs and charts and summaries but how can we turn that into advice that is actionable and highly personalized and tailored to that individual? And that's where generative AI helped us translate all this into those personalized recommendations. That's a great call out there. I don't want to get Bhaskar's opinion on this because I think that personalization angle is huge. That's where value comes in because now it's just you're tailoring the data for individuals yet leveraging all the corpus of data. So the quality is super important. This is a really big deal. Yeah, I mean from day one, accuracy and the quality of the data has been one of the number one tenants for us. The data combat abilities between the data that we're collecting from a youth player to the MLB player. We're making sure that there's no data fragmentation between all those levels of play. But the other part that's really important and we've been working with persistence but we have a full sports science staff that's been looking at our data over many years and then been working with the professional teams as well as youth players. And so that journey for a youth player is gonna be different than what the journey is for a high school player or a college player. I mean, the goal is always to get better but it's not necessarily working on the same metrics at each level of play. And one of the big initiatives for us for next year is to focus on the youth and gamify our product offerings. And so make it fun but while you're fun and playing you're actually getting better. And so part of the goals of BLAST IQ is to be able to have different personalities depending on the level of play. Whether it's girl softball players or guys want to play baseball or golfers the personality of IQ is going to change. The personalization angle and also the value of the data, great point there. As you look at the challenges I mean a lot of upfront workers this is a great example of what it takes for real business use cases. Everyone sees in the AI hype today, co-pilot, write code, actually augmenting the human aspect is really key. This is an example of that. The human plus the AI helps the player from youth to pro. What are some of the challenges here? And from your standpoint as a business model and also from persistence standpoint getting it going, you mentioned wrangling was the adoption curve fast? Where did it start to accelerate? Take us through some of the early launch of the products you guys did. How long did it take? What was the front end work? What's the operating cycle look like? Yeah, sure, absolutely. So BLAST Motion was already on a journey to AWS cloud and given AWS is extensive not only the data architectures and services but AI services and then generative AI services. It was a natural choice for us to go with that as well as the kind of agility it brought the kind of ability to scale from a POC to production. So it made a natural choice. So we've gone with a slew of these tech we've using AWS SageMaker for the initial data wrangling establishing baseline. BLAST Motion is already using Aurora DB and DynamoDB from AWS to store some of their data. We then used LangChain to tie all of these together and bring out the LLM context in there. We used Bedrock and the Titan model specifically but as well as I believe Jurassic model from AI 21 Labs was also used for some of the LLM related tasks. So this was kind of broadly the stack that brought this to life. And this is going to evolve over time as the more services become available on the hyperscalers as well as our needs for the kind of outcomes we want to achieve for the players. Eventually that's where our North Star is and we work backwards towards technology from there. That's, hopefully that answers your question. I think that's a great example. I just want to make a point if you don't mind. I think when I talk to Matt Garmin who runs the field global services for AWS he used to run EC2, he knows this stuff. He was talking about how partners had access to Bedrock. I believe you guys had early access and that was kind of key for your team here. Is that true? Did you guys have early access to Bedrock and how did that help you from a partner angle with AWS help your customer? Because this is an example of getting in early and having the right people around the problem statement. Absolutely, yeah. So we have given our multi-year relationship with AWS both as a partner who builds solutions on AWS and as a consumer and as an engineering partner. We are able to get early access and bring that to bear in the particular blast motion engagement. And that's been pivotal because the data that blast motion has not only is it sensitive from the players that they collect it from but also it is the unique mode that they have for their business. So that the confidentiality of data was paramount and the security of data was paramount for both them and for us. And so having Bedrock Early Access allowed us to build this out, experiment with it and also build it out rapidly without having to worry about data confidentiality, data security, because all those controls are available de facto within the offerings from AWS. Interesting, those could be roadblocks. You guys turn them into an opportunity. Absolutely. With blast motion and Baskar, this is an example of how the business transformation of cloud, next-gen cloud changes your product. You can get faster, better, more personalized, scaling up more capabilities as you get more data. The flywheel starts going. How are you looking at this from your business? Because you also want to make sure you've got the guardrails in place to making sure that you're protecting your IP and the data is going to continue to feed in. Share your perspective on how you're positioned. Yeah, I think security and accuracy has always been one of the most important factors. So a lot of this analysis, we make sure that the data is anonymized so there's no personal information that's being used in any of these kinds of analysis, but we do know level of play. So we know a college player versus a high school player versus a travel ball team, whether they're a professional golf pro player, we have those categories, but it's all anonymized. So we're very careful on that relationship. The fact that Persistent had early access to bedrock and Titan, that is huge for us because as we enter into this space, we're looking to them for that guidance and you have to understand, it's not like Persistent just brings a few developers to the table and then starts implementing AI. They brought their advanced R&D team, consists of data scientists, AI experts. They're bringing the full breadth of everything that they have learned with early access to bedrock and then leveraging our data to basically provide a solution. So this is a very strategic partnership for us. I think the support from AWS, understanding that they have to also provide us early access to bedrock. So the AWS team here, that's our account manager, they went to bat and they made sure that we also had early access to bedrock. So all these things play together and it puts us in the forefront because when we release to this market, we'll be ready with an advanced solution that's been vetted. That's awesome. And I guess the key question I want to know is when you see those early talent stars rising, is there going to be a talent agency kind of forum, that nice little opportunity to get in early and sign those sports mega stars before they become huge, almost predictive. Yeah, I don't want to say anything about that. Pender, a great example of persistent again. This is where you guys are at now. I guess the next question to wrap this up is that where do you go from here? Cause one of the things that's clear in our conversations on theCUBE is everyone's moving faster than ever before. It's an exciting time. You got the guardrails, you got the advanced team. This is just the beginning, it's a start line. As this new Genervai comes in, time to value, product market fit, these are going to be accelerated. What is your view here and talk about that trend of what's next with you guys as partners? Absolutely. So this is like you said, just the beginning, right? The technology itself is going to mature. The ability to bring this technology to multiple different user experiences is going to get even better as time passes and it's getting there fast. But I want to be clear about one thing, right? Generative AI just like any other technology is a means to an end. Our mission with Blast Motion is absolutely to make sure that these players, whether they are a kid doing this on a weekend or it's a professional player, they are being helped to become the best versions of themselves and our mission is to make sure Blast Motion succeeds in that. So with that as a North Star, we are looking forward to not only bring more user experiences, but to scale this out to new heights with the cloud infrastructure and the modernization journey that we are on with them. Well, I certainly want to participate if you need any extra subjects to be part of the beta or testing, happy to join in. Bhaskar, I'll give you the final word. What are you most excited about going forward for your business as you embark on this endeavor, as you go this next level? What are you most excited about? Well, I'm excited about the partnership between AWS, Persistent and Blast. It's definitely, you know, it's something that we're able to leverage as we move into the future. And the beauty of that partnership is that it allows us to focus on what we do best. You know, one of the concerns that people have is that technologies like generative AI are going to replace, you know, people's jobs. We have a pretty strong biomechanic staff, our sports science staff, but now they have way more work than they ever had before because they can focus on the player journey. And what I'm really excited about is the player journeys for youth because that, besides being the widest part of the market, it is also where we can make a difference. When we come across the, you know, professional players or, you know, the college players, the elite college players, they're pretty much set in their mechanics. But it's where we can make the biggest difference about, you know, at the youth level. That's what I'm really excited about. Well, congratulations, Penderon. Congratulations on persistence. You have the right team. Thank you. Great customer blast motion. I'm a fan. Great conversation. Two doctors, their PhD, Dr. Baskar Bose, and Penderon, come on. Thanks for coming on theCUBE conversation here. Business transformation, generative AI is the next big wave. It's happening fast. Get your data, protect it. It's a competitive advantage, and new use cases are emerging and it's highly accelerated. Of course, it's theCUBE, bringing you all the action. Thanks for watching.