This discussion at NYSE Wired for theCUBE AI Factory series examines enterprise artificial intelligence, hereafter AI, sovereignty and agent orchestration for large organizations. Trevor Hansen of Blunom.ai, founder and CEO; Serge Shevchenko of Blunom.ai, co-founder and head of revenue; and Brandon Kissinger of SnapSoft, executive chairman, join theCUBE hosts John Furrier and Dave Vellante to explore building a sovereign AI control plane and orchestrating agents for enterprise use.
The conversation draws on theCUBE Research and AWS experience to address agentic workflows, integrations with 150+ independent software vendors, hereafter ISVs, model and hardware modularity, cost optimization and secure deployments across on-premises and cloud environments.
Hansen emphasizes the need for organizations to take back control with a sovereign AI strategy and to implement a centralized control plane for governance and policy enforcement. Shevchenko recommends preventing shadow AI through centralized orchestration and standardized integrations; they highlight integration patterns that reduce risk while accelerating deployment. Kissinger emphasizes partner-led, outcome-focused rollouts that target high-return agentic workflows and deliver measurable profit and loss impact. The panel stresses model swapping, hardware optimization and continuous cost management as operational imperatives for enterprise AI.
This session provides practical guidance for technology leaders and enterprise IT teams on implementing sovereign AI strategies, orchestrating agents, optimizing total cost of ownership and securing deployments across on-premises and cloud environments.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: AI Factories - Data Centers of the Future. If you don’t think you received an email check your
spam folder.
Sign in to AI Factories - Data Centers of the Future.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open the link to automatically sign into the site.
Register for AI Factories - Data Centers of the Future
Please fill out the information below. You will receive an email with a verification link confirming your registration. Click the link to automatically sign into the site.
You’re almost there!
We just sent you a verification email. Please click the verification button in the email. Once your email address is verified, you will have full access to all event content for AI Factories - Data Centers of the Future.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
OR
Connect via Twitter
Connect via Linkedin
EDIT PASSWORD
Share
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
theCUBE + NYSE Wired: AI Factories - Data Centers of the Future. If you don’t think you received an email check your
spam folder.
Sign in to AI Factories - Data Centers of the Future.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open the link to automatically sign into the site.
Sign in to gain access to theCUBE + NYSE Wired: AI Factories - Data Centers of the Future
Please sign in with LinkedIn to continue to theCUBE + NYSE Wired: AI Factories - Data Centers of the Future. Signing in with LinkedIn ensures a professional environment.
This discussion at NYSE Wired for theCUBE AI Factory series examines enterprise
artificial intelligence, hereafter AI, sovereignty and agent orchestration for
large organizations. Trevor Hansen of Blunom.ai, founder and CEO; Serge
Shevchenko of Blunom.ai, co-founder and head of revenue; and Brandon Kissinger
of SnapSoft, executive chairman, join theCUBE hosts John Furrier and Dave
Vellante to explore building a sovereign AI control plane and orchestrating
agents for enterprise use. The conversation draws on theCUBE Research and AWS
experience to address agentic workflows, integrations with 150+ independent
software vendors, hereafter ISVs, model and hardware modularity, cost
optimization and secure deployments across on-premises and cloud environments.
Hansen emphasizes the need for organizations to take back control with a
sovereign AI strategy and to implement a centralized control plane for
governance and policy enforcement. Shevchenko recommends preventing shadow AI
through centralized orchestration and standardized integrations; they highlight
integration patterns that reduce risk while accelerating deployment. Kissinger
emphasizes partner-led, outcome-focused rollouts that target high-return agentic
workflows and deliver measurable profit and loss impact. The panel stresses
model swapping, hardware optimization and continuous cost management as
operational imperatives for enterprise AI. This session provides practical
guidance for technology leaders and enterprise IT teams on implementing
sovereign AI strategies, orchestrating agents, optimizing total cost of
ownership and securing deployments across on-premises and cloud environments.
>> Palo Alto Studio connecting Silicon Valley and Wall Street. I'm John Furrier, with Dave Vellante, my co-host. Welcome back to theCUBE here. I'm John Furrier, your host here at our New York Stock Exchange CUBE Studios, part of our East Coast hub, of course, connecting Silicon Valley from Palo Alto to Wall Street. It's our AI Factory series, part of our NYSE Wired program at CUBE Original. Got two great friends on the CUBE. Former Amazonians now starting a new venture. Again, coming out of stealth, kind of hitting the agent market. We got Trevor Hansen, founder and CEO. Blunom.ai. We'll get into the name in a second. Serge Shevchenko, who's the co-founder and head of revenue. The pressure's on you. Serge, great to see you.
Serge Shevchenko
>> Yeah, likewise.
John Furrier
>> Trevor, congratulations. This is your first public launch video here in theCUBE. Appreciate you giving us access. We've known each other from the Amazon day, so congratulations.
Trevor Hansen
>> Yeah, thanks, John. Firstly, it's been a tremendous journey. We've partnered multiple times throughout the journey, seeing customers go through the cloud migration now trying to figure out how to implement AI. I spent a lot of time deeply understanding the problem statements across companies trying to securely adopt this, and no better way to solve it to go build the product myself. So, being really passionate about building this sovereign AI strategy for customers. At a very high level, we realized that enterprise customers and every persona is sitting there trying to figure out how do they invest in this technology? How do they keep it safe? And then more importantly, how do they standardize? We came to realize that having a sovereign strategy is critical so that you can actually understand where you're deploying your stack, your datas, your own data. The model providers you're using, you have controls and boundaries around that. And that's what we set out to do is ensure we can equip customers with that technology.
John Furrier
>> Serge, talk about the unique Amazon position, because Amazon sees everything. You mentioned sovereignty. It's the hottest areas where the most of the infrastructure build out is where the cloud intersects. I mean, cloud sovereign has been around for a while, but now AI sovereignty's coming in. It speaks that you guys had a good visibility. What did that enable you to do, and how have your combined experience collectively bring you to this opportunity?
Serge Shevchenko
>> Yeah, I mean, I think spending time with partners and customers over about a decade. So, collectively we've been doing this for about 20 years. We noticed that the AI bottleneck was not just services or security or cost management. You're seeing some of the articles about organizations facing these challenges. That the challenge is actually the lack of the combination of both, proper AI orchestration and services, to really approach this new software development model like we talked about.
John Furrier
>> Trevor, you've seen a lot of the corp dev, product work, partnerships. Amazon Web Services probably was probably the best success story that I've seen in my lifetime. Now we've got the whole AI thing playing out too. How it's gone from just such a humble beginning, misunderstood as Andy Jassy would say, to this massive ecosystem, major enterprise penetration, just the growth. And okay, that hits it. Successes are there. Then you got this whole, that's the substrate now we're building on. Now you've got the agents coming over the top that's now globally distributed. What is unique about this venture? What jumped out at you? What made you guys start the company?
Trevor Hansen
>> Yeah. I mean, the first is partners play a critical role in this journey. So, there's existing technology providers, ISVs and consulting companies that are helping these customers migrate, modernize, unlock that value. The problem is when you're implementing this technology in isolation, there's a couple of problems that come up in the enterprise. You have shadow IT popping up in every line of business, because they're all procuring AI software individually, and with good intent.
John Furrier
>> Channel AI or IT?
Trevor Hansen
>> Well, both.
John Furrier
>> Both.
Trevor Hansen
>> It's both, because this technology-
John Furrier
>> IT has become the cloud, basically.
Trevor Hansen
>> Yeah. This technology isn't isolated. It actually integrates with your data and you need to make sure that you've got the right security protocols in place so that the agent doesn't index your whole database and spin up thousands of dollars of unnecessary cost, let alone pulling IP into a model provider where you want to expose your company. So, ensuring that the company has the right control plane in place and they have the controls in house allows them to be effective at evolving the business from a traditional architecture to be agentic. And when we started building this, we realized that partnering early for success was critical as well. So, working with consulting companies that give them the ability to choose their model provider, choose their cloud and choose the outcomes they're trying to invest in.
Serge Shevchenko
>> Yeah. And you talk about AI factories often here, and I think we're also being mindful about the infrastructure and the hardware layer as well, ensuring that agents are building on the right hardware and the most efficient hardware as possible.
John Furrier
>> On the AI factories, how do you plug into that? Because one of the pain points you were just mentioning, Trevor, is that the enterprise adoption is chaotic and it's total chaos, not even managed chaos. I mean, cloud, shadow IT was managed chaos, because essentially I was basically going around bottleneck blockers and organizations and getting a better deal in the cloud. That was good chaos, but it was more of a IT disruption. Shadow AI is just totally chaotic. Things can run wild, grab credentials, and the databases are involved. I mean, it's essentially a data sprawl, uncontrollable.
Trevor Hansen
>> Yeah. And the bigger risk for leaders at companies is they actually don't even have visibility into this. What agents are accessing, what data at what cost in their business. And when they sitting there saying, where are we deploying AI technology today? I'm pretty sure most CIOs and COOs are not able to actually show you a clean list. The risk starts all the way in the existing data centers where they're busy deploying technology to harness existing data. What we've done is actually looked at every layer of the stack and we partnering with infrastructure providers so that we can do cost optimized deployments within a customer's data center, so that when they're going through migration and modernization, they have a sovereign AI control plane that they can migrate and over time harness powers in the data center and in the cloud.
John Furrier
>> Go ahead.
Serge Shevchenko
>> Enterprises want the ability to fully control AI, not just in the cloud, but in the data center as well. And we give them the ability to do that, and the portability that when they are ready to go to the cloud, they can do that.
John Furrier
>> So Blunom, B-L-U-N-O-M.ai is the name of the company, Blunom. Serge, you mentioned this kind of like a genome. My son just did a genome sequencing and you see all the DNA and genetics. As mutations happen in the DNA going back to the origination. Why the name? Explain the name Blunom.
Serge Shevchenko
>> That's a great question. I don't know if you saw the new Fed chair, Kevin Warsh, just talked about how AI will just be called business. And we believe that through our platform, as we provide every line of business leader the ability to build, manage and deploy AI safely, securely, manage costs, we are truly like genome to the body, we are going to bring businesses to life.
John Furrier
>> And there's a lot of DNA in there.
Serge Shevchenko
>> Yeah, and to have-
John Furrier
>> That's an Oracle database from them, but system of record. Don't touch it. It's all our Salesforce stuff. So all these databases, this is why SaaS apocalypse is totally overblown, because that data gravity and mode, you can abstract that away. Now, they got to build better products. It's like the iPhone versus the iPod. They're two different things.
Trevor Hansen
>> Yeah. And to add to the Blunom, so blue being a color you trust from traditional IT, but the nom is the genomic aspect of your business. You need an antivirus system that the nervous system that's connected. One thing that we're really excited about is having a cognitive layer that allows the business to centralize knowledge and learnings. You talk about SaaS. We have over 150 different ISVs integrated into the platform that the agents natively understand how to interact with. When you're building an orchestration layer asking a simple question of your AI agents, that agent is able to actually go interact with securely with different software companies where you already have business users operating today. The valuable part is that you're able to deploy this technology to both technical and business users. We see a lot of companies today, they're building specifically for the technical persona and the developers. We believe that equipping every line of business, particularly the business users with a piece of experience in the AI that can drastically improve their productivity is going to be critical, but you have to do it with the right controls in place.
Serge Shevchenko
>> That's right. Well, and also just to touch on that, giving the CISO the ability to manage these agents, regardless of which line of business they're being deployed in. That's some of the challenges our customers have been telling us. They're seeing a lot of these agents being managed, built, deployed in different lines of business, and they have no idea how to manage not just cost but intra-agent communication or department communication east to west, which we're solving for today.
John Furrier
>> It's an integrated approach. The C-suites involve the line of business you mentioned. We're seeing that in the data from theCUBE and theCUBE research. It's not just an IT transformation, it's a business model transformation, it's a complete thing. You mentioned the partnerships. Knowing you guys were in stealth for a while, I know you were very kind of low key and I didn't even know what you were working on until I just found out today, so thanks for sharing. But how much work has gone in? Take us through from origination, how much momentum have you did? What work have you done integrating 150 ISVs?
Serge Shevchenko
>> Yep.
John Furrier
>> That's pretty significant. We're going to have Brandon Kissinger on, who's one of your first customers. You've done some work.
Serge Shevchenko
>> And partners.
John Furrier
>> Yeah. So take us through, because you got a lot of traction there. You kind of built it out. Take us through that.
Trevor Hansen
>> Yeah. I mean, what I saw firsthand is no enterprise customer is able to solve this problem alone. And they're bringing in specialized technical individuals. They're bringing in consulting firms they trust, so multi-partner collaboration to unlock value is going to be critical.
John Furrier
>> You're building an ecosystem stack basically into your layer?
Trevor Hansen
>> And that's something we're proud about is that we ecosystem orchestrators as well is that the channel from the beginning was a priority at the top of the board for us is realizing that we should not just go in alone, but partnering with software companies that deeply understand AI cloud migration modernization.
John Furrier
>> All right, so here's a question for you. In the training to inference evolution, everyone was one point into training, buy a bunch of GPUs, train the heck out of the data and then comes interested away. I bought that rack and I got to redo it. We're seeing a similar thing play out for agents. I think I nailed my layer and then all of a sudden, wait a minute, I need harmonization. I need compliance. Are you seeing the same thing play out? And what should customers do? Because if I could swap out models, I need to create some fusion.
Trevor Hansen
>> Yeah. I mean, now a couple of years into adopting AI and agents, you now realize that the infrastructure layer is going to be the most important aspect of this deployment is ensuring that this cross-functional collaboration across your business, that the models can communicate with each other. And more importantly, you're not creating data silos or AI silos in your business. So, having a platform that can connect the whole business and every agent can speak South to West across the business.
John Furrier
>> And Serge, talk about the model piece, because what you're seeing is small language runs, which we called four years ago, so we got that right. But now we're seeing real time swapping out models because of either economics or capabilities. It's like that scene in The Matrix. Upload how to fly a helicopter. You might need a model today for this.
Serge Shevchenko
>> Yeah. If agents are airplanes, and we consider ourselves as the FAA, right? We're making sure that people are not shipping a box from one city to another or perhaps even one neighborhood to another using a 737. You don't need to do that, swap it to a local model. Or perhaps train your own model and have the ability to sort of swap models based on the user need.
John Furrier
>> So, you're building intelligence into the layer.
Serge Shevchenko
>> Very much so.
Trevor Hansen
>> And to put a pin on that is, the more the platform gets consumed within their own environments, consumption makes recommendations. So, it actually starts recommending that the model you're using is overkill for email summarization and you could actually improve your cost by 70% in token usage, and it makes the recommendations to the business to swap out a model.
Serge Shevchenko
>> And add another layer to that, because model modularity is important, but add another layer, which is model modularity and hardware, suddenly you're looking at significant cost savings on the customer side.
John Furrier
>> All right, guys. So talk about the status of the venture, financing, headcount, what's your goals you guys hiring? What's the plans? Give a little stats on inside the numbers, what's going on or what's going on with the company.
Trevor Hansen
>> Yeah. I mean, we first set out to make sure we deeply understand the problem statement, build the right solutions, the right partnerships and engage in true business outcomes selling. Our goal is not to just go when drop in a product and step back. We really want to make sure we have success at the front of our partnerships and our customers. So, our first call was to a partner, and we're really proud that we built that out first. We're actively in conversations with investors and we're excited about those, but we're in the long run, we want to make sure we create compounding value.
John Furrier
>> So you have not done a big VC round yet?
Trevor Hansen
>> We're actively in conversations with venture capital and investors, and it's going to be a key part of our growth story, but customers and partners will always be our first points of entry.
Serge Shevchenko
>> Correct.
John Furrier
>> Great guys. Congratulations. Trevor, we just had a great session with Serge. Now Brandon Kissinger, the executive chairman of SnapSoft, your early design partner, and I love that you're on, Brandon. Thanks for coming on theCUBE and participating. A startup success is make or break by their entry strategy, right? You win or lose by your risk management, but it's a gut feel. So, smart money always gets a nice partner, design partner they call it. That's what you guys were. Explain the relationship that you guys have and how'd that come together?
Brandon Kissinger
>> Yeah, absolutely. And John, thanks for having me. I ended up sitting down with Serge over a cup of coffee. I told me about an idea him and a colleague had, started as going to kick off and we deploy agentic agents in the AWS ecosystem all the time. We're a premier tier AWS partner, been around since 2016. And what's always fascinating to me is you got to take a stance on this and truly believe that when you deploy agentic agents, there has to be a control plane. There has to be governance. You have to think about token economics if this thing is going to scale. And I was blown away when I sat down with these two and got to see the platform firsthand, and we've adopted it internally and for all of our customers.
John Furrier
>> And Trevor, you were detailing out with Serge about the vision and we talked about you guys pedigree at AWS. Now you have an AWS partner tier one. When you look at that, the cloud game was hard. If you look at like the growth of cloud, it was over a decade and a half or more than how you look at it. But now the AI thing's shooting up even faster, built on cloud native. I guess my question for you guys is, as you see the market surge up with the AI, you still got to run in cloud, but the on-prem piece has token economics built in there, because all AI teams will get built on premises first, almost like a land environment given the cost of tokens. So, you're starting to see actual enterprise workers get enabled for the first time using the tools as a utility and they're burning through their token budgets. So you're going to see a ton of on prem where the data is, but the cloud's not getting any smaller either. And now you've got international with sovereign cloud. So now there's a whole nother architectural reset. So the question I have for you guys, how do you see that architectural reset? Because it's not, that throw away the old, it's building on top of. How do you guys fit into that narrative?
Trevor Hansen
>> I mean, I can add to start. So I believe meeting a customer where they are in their journey of from their data center to the cloud. So they've invested significantly in existing infrastructure applications, data that they house internally that you don't have to wait to unlock that value in the cloud out the gates. What we have built as an orchestration platform that can drop into the data center with our infrastructure partners. And the key part there is partnering with services partners that can help you understand your current estates in your data center and help you be able to surgically identify workloads to move to scale into the cloud. And that's where the Better Together partnership comes in is having a technology platform that is sovereign and a services company that you trust that can help you through this transformation.
John Furrier
>> Brandon, you've seen a lot of the lift and shift, now the microservices come out, higher level services from Amazon. As the AI comes in, how's that intersecting from your standpoint?
Brandon Kissinger
>> I'll tell you the free cash flow that it opens up for a business is where I see so much of the importance when it comes to AI, and especially Agentic AI and its workflows, especially when you can start to operationalize it at scale, which is again what the Blunom platform helps us to do. But whether the customer's in the data center, whether they're in the cloud, I see this huge opportunity to migrate payroll costs into cloud consumption, usually at a discount 60, 70% of what you would have spent otherwise. And it's fascinating to see the level of human productivity that goes up in the company. And I'm sure like every business owner, I got folks whose sole job is to put revenue money at the top of the funnel, and then I have whole teams of people who need to bring it down to the bottom. And the more of it that gets at the bottom, the more I get to reinvest in growing the company. And that's the story I talk to with customers for why they adopt and go down this agentic journey.
John Furrier
>> Expand on that, because this is where the revenue action's coming in. If you look at all the agent successes I just did interview this morning with a founder of a company that's been around during even the big data days, and he's in the governance area, he's crushing it. But the revenue is really, for the ones that do agents right, see revenue, not just cost takeout. And now you bring up the other issue of, where's the budget going to come from to buy more tokens or build more teams? So, the economics become big and I really don't like the word tokenomics because that's just cost of tokens. There's other costs in the business model. Could you share your thoughts on how you see that playing out, because you were hitting on some of those things where, "Okay, yeah, I got, but I still got to reinvest."
Brandon Kissinger
>> Spot on, spot on. And I have feelings about it, And it's one of those things, there's a bit of a Groundhog Day. I was at the AWS Partner Summit sitting around the room with a bunch of guys and gals just like me who run SI partners, some ISVs, and it was fascinating. They got asked the question, "What would it take for you to implement more agentic AI in your customers and in your own company?" And without question, everybody came up well that we got to clean up the data first. And what's fascinating, because this is a change over the last few months, you don't have to clean up the data anymore. We're building agentic agents that actually complete a workflow. And when I say agentic agents, I'm not talking a singular agentic agent, I'm talking we're building full stack agentic applications that finish a process end to end. And so those token economics become critical when you're looking at that actual activity that person is going to do. Sometimes accuracy has to be higher. Sometimes it can be lower. You can cost save here, you can cost save there. Sometimes you have to spend a little bit more. You have to make sure that if you build out an agentic workflow to do a creative type of project, like write an email or a paper that they're not submitting it 47 times to change one sentence.
John Furrier
>> I mean, the cloud game was great. Dave Vellante and I would always joke about pay by the drink, USC pricing. When you get to the agents, they're drinking a lot. So the token thing, I mean, all kidding aside, this is policy routing, this is service level agreement like thinking. This is like an OS, it's got scheduling involved. I mean, this is a complex system.
Trevor Hansen
>> To add to that quickly. I think as leaders at a business, you want to see P&L impact. Your orchestration layer shouldn't just be executing the workflows and keeping those agents secure, but more importantly, it should be surfacing up recommendations for efficiency. And that is purpose-built workflows, more deterministic, what model providers should be plugging into your agents and you should be seeing that efficiency gain. So as your token spikes, essentially you should be seeing it continuously optimized. On the top line, that's where services partners come in is directly partner with the CEO and say, "What are the key priorities for the business? Where should we be investing the efficiencies we've gained?"
John Furrier
>> I think you guys are hitting an organizational kind of like a Venn diagram, everything's coming together, the C-suite's changing their roles, the engineering teams, partnerships. And then at the middle is the AI builder. I call it the AI builder persona, because it's not just a developer, it could be a business person. In the C-suite, the line of business owner's elevated, and now the C-suite's got the CFO and HR involved, CSO and CIO are super positions.
Brandon Kissinger
>> Absolutely.
John Furrier
>> And then the platform engineering and the data science, they're all coming together. So, in the middle is the AI builder. If you believe that, you say, "Okay, that's going to be the killer persona." How do you see that operating model working, and what attracted you to Blunom? Because that's the optimization point, because if I'm a builder optimizing payroll or closing the books in the C-suite or running platform engineering, I'm doing, I'm the builder.
Brandon Kissinger
>> So putting in these agents all throughout your company is like hiring hundreds of interns and giving them access to all your critical systems. These agentic workflows are not George Foreman grills. You can't just set them and forget them. And that's what attracted us to the Blunom platform the second I took a look at it. And the orchestration layer and the visibility it provides to my engineers as we build, as well as our customers when we deploy it and manage those is phenomenal. And every business does this. Every year they sit around and they do forward planning for the next year and there's KPIs that all of the C-suite, those line of business owners decide to hit. And then they ask for all the budget and headcount to be able to hit those KPIs. We did the same thing internally at SnapSoft. And then I sat down with the team and figured out which of the future headcount could we replace with agentic workflows. And it was a fascinating process to be able to show the ROI that way.
John Furrier
>> How long was the list? Must have been a huge list. And then let's take these three.
Brandon Kissinger
>> If you think of Snap, we came up with 16 Agentic workflows across my LOBs just off the top of their head, and then we started to build it out. But there was a key thing we learned in that process is that somebody's job has to be to manage those agents and those workflows full-time. It could be internal at the company if you have it, or you can outsource it to a partner like us to be able to manage them, build them, identify and build them and manage them for you. I sort of joke the CFO and the COO are my two biggest supporters at the customers we talk to, because those are the ones who always have-
John Furrier
>> Well, they can see the shadow AI hit their department. That's why shadow IT really drove the cloud business because the guys knew, "Hey, I put my credit card down, I'll get the prototype up, then I'll prove it and then I'll get budget." Now that's happening in every department like, "Oh my God, this is actually real."
Trevor Hansen
>> I want to come back to your question, which is, the reason why all these personas are gravitating towards each other to say, "How do we solve this together?" To me as an indication that businesses are going through a big transformation. They're trying to figure out how as a leadership team, they're going to invest in change and securely adopt AI. So when we think about our platform and the way we built it, we thought very meticulously about the CEO's AI strategy, the CFO and how we give them financial control, the COO on how to create orchestration and essentially a workflow management platform that could deploy to every line of business user, not just technical, but empowering every business user to be efficient.
John Furrier
>> Okay. So Brandon and Trevor, talk about where you guys intersect and what that means for people watching. Are you in business today? How do people engage? What's the status of the go-to market? What are you looking to add from a tech stack perspective?
Trevor Hansen
>> Yeah. So June 9th we're going into public preview, so please reach out to us, Blunom, also our partner, SnapSoft. But I would leave every leader watching this with three key takeaways. The first is take back control of your AI strategy. You have to think about a sovereign AI strategy for the long term. And when I say sovereign, I'm saying, bring back some control in house so that you have the controls, the levers, and you can see a future with the right platform. The second is partner early for success. You have to bring in specialized skilled people that share the same outcome interests that you have that are able to help you meet you where you are and essentially transform. And then finally is empower and democratize every person across the business. Stop focusing just on developers in the IT department. Your business users in the line of business are all trying to find technology to help them be more productive. The problem is they're all buying off-the-shelf vertical solutions. They're creating shadow IT for you and also limiting your visibility. So standardize on a control plane that empowers you.
John Furrier
>> Yeah. I'll say get some managed chaos, not full chaos. Brandon, talk about the execution of all this, because at the end of the day you got to execute. I mean, I can buy that strategy all day long. Sign me up. I want to take that hill, put AI and everything in my business. But to get to a system of execution, let's talk about sovereignty is hard. Talk about what you guys are doing and how this fits in as you guys execute, because you guys are also transitioning in real time too.
Brandon Kissinger
>> Yeah. John, that's the right question to ask. We are an all-in AWS partner and they have phenomenal programs to be able to make this easy to implement for customers. If you think about the three-legged stool for why somebody does or doesn't do something, expertise in house, the budget, and then of course the bandwidth. Not everybody has employees sitting around not doing anything ready to experiment. I'll tell you the way we've done it and the way we've packaged it up is if you followed me down, my KPI to budget and headcount conversation, we go into a customer, we talk to the different lines of businesses. We look to identify one agentic workflow end to end with the highest rate of return for the company, but the easiest to implement. And this is going to involve us building multiple agents with multiple integrations. Blunom does a phenomenal job with the integrations. And then we build that Agentic workflow in 30 to 45 days soup to nuts and it's ready for UAT, user accepted testing. We give it to the teams to play with it and give us feedback. Then we run you all the way back and then I say, "John, we did one for you. The team is happy. They're using it. Give me, let's do the next four to six." And we're going to run it through over the next two.
John Furrier
>> And you guys are intentional at the targeting of that workload.
Brandon Kissinger
>> Yes.
John Furrier
>> This isn't a risk management side project. This is let's target where revenue business value is hitting.
Brandon Kissinger
>> And we need to figure out, we need to decide on what that ROI is at the very beginning. I'll tell you internally, we have our AP and AR process automated with agentic workflows, our allocation process with agentic workflows. So many of our business unit and teams enabled with agentic workflows and the ROI becomes clear. But get your first one, I would tell everybody, get your first one done and then we'll build out a half a dozen for you.
John Furrier
>> Do you see the kind of makeup of an individual ... Let me rephrase. Is there a threshold for implementing that kind of full stack? When does someone qualify to and leverage a platform, or how does someone become AI-native with it? What's the entry point? I guess that's my question. How would you guys talk about it?
Brandon Kissinger
>> Trevor might have an opinion on this one, and it depends on the workflow. For the one I'm talking about specifically, SnapSoft is a few hundred employees. I have two dozen agentic workflows that I can deploy, and it's the ones who ... The companies that do get enabled by AI are the ones who are going to have more free cash.
John Furrier
>> It doesn't have to be a big enterprise.
Brandon Kissinger
>> It doesn't have to be. I think even a lot of the .
John Furrier
>> SMB.
Brandon Kissinger
>> Absolutely.
John Furrier
>> Cloud-native company, obviously with Amazon.
Trevor Hansen
>> I think to add is what's nice about our platform is you don't have to have enterprise capability deployed out the box for an SMB. SMBs don't have a technical or an AI department.
John Furrier
>> So, we have DevOps at theCUBE. We have full AWS stack. For us, we could leverage it.
Trevor Hansen
>> Yeah, exactly.
John Furrier
>> We have workflows, not a lot.
Trevor Hansen
>> Yeah. So, when you look at customers that are trying to implement this, digital native are going to be moving a lot faster. You have 500 to 2,000 users. They're the ones leaning and investing in this technology. Ideally, they've been building some version of a stack in house, and they realize they better serve moving those engineers to customer-facing experiences versus trying to build their own orchestration plan. So I would recommend, if you are sitting there thinking about your economics and the amounts of developers that you're building in house, what I see is, it's really compelling when your developers moving at 10 times the speed in the past, but the thing is, what are they working on as an outcome? Are they actually building a product that's going to impact your P&L? If they're not, bring in Blunom, bring in SnapSoft. We'll dive in and help you identify use cases.
John Furrier
>> Use I think the P&L impact's huge. I guess my final question for you guys, because you've got great visibility, I'm curious. There's a huge set of customers out there that are on AWS and have had all the goodness of the cloud doing great. I mean, they're building platforms on top of AWS. Now they have to architect the journey. What's your advice for people? Because I'm imagining with all the inside baseball you guys have of AWS, there's a lot of nice intersections between the on prem sovereignty global footprint, dealing with regions, dealing with fail back over, data privacy, revenue generation in country. What's your advice for customers out there? Amazon Web Services customers and other clouds.
Trevor Hansen
>> To add, and then I'll hand to Brandon, you need to lean in with your AWS account management team and tell them your long-term strategy and call out partnerships that are critical to you. And that's something we've really enjoyed is that the account teams are partner friendly. They are co-selling to help unlock cloud value, and there's programs that help with migration, modernization, AI adoption, AI assessments, et cetera. So really making sure you're leveraging the channel of AWS. So for our engagements for every customer, we're going to contact your AWS account manager, say, "Hey, this is us. We're partner obsessed. We're coming to help you unlock your cloud journey with this customer, and we're going to be in the accounts for the long term."
John Furrier
>> I guess my question is, where do I plug in to AWS? I got my partner management, but I'm thinking of architect. Do I just plug into the stack?
Trevor Hansen
>> And that's where our business outcome orchestration part of our businesses is we immediately dial and connect up all the different stakeholders together. So, if you don't know the right place to go, contact us and we'll help you-
John Furrier
>> Can I do self-service?
Trevor Hansen
>> Yes. There's a self-service you can authenticate with Microsoft or Google on the platform and your users can populate, but at the end of the day, you really want to make sure that you're transforming with partners for success. So as much as you can make your own progress, which the platform provides and a SaaS-like experience as well.
John Furrier
>> So will you guys be in the marketplace?
Trevor Hansen
>> Yes. AWS marketplace is going to be a key point here is really helping those enterprise customers draw down their investments in-
John Furrier
>> So, I could consume through the marketplace?
Trevor Hansen
>> Correct. So marketplace private pricing agreements or just standard subscription is going to be a key part, and we're excited about-
John Furrier
>> And how are you guys thinking about revenue model just while I got you here?
Trevor Hansen
>> So revenue, and that's why we're moving away from per seat licensing. We believe we're going to be pretty disruptive in the economics of SaaS and software. We looking at platform and outcome based pricing. So really making sure that we are pricing our model for the long term and seeing customers invest more in us because they see results.
Brandon Kissinger
>> The more customers I onboard to it, the more they charge me.
John Furrier
>> So Brandon, I got to ask you a question since you're here, because I'm fascinated by the ecosystem evolution. How do you see the cloud ecosystem evolving to this AI-native ecosystem? Because the game is still the same, but the game has kind of changed. APIs are great, but now you've got agent connections, you got a lot more co-design going on, a lot more engineering in the partnerships. How do you see the success formula in the ecosystem to be successful building on top of cloud and to take advantage of the AI native?
Brandon Kissinger
>> Oh, well, I'll talk about it maybe from a personal standpoint as an SI partner. And I think this is where a lot of companies have to wake up and get real that there's going to be parts of your business and part of the things you do that you're going to have to cannibalize, or somebody's going to eat your lunch. And we looked at it internally even for us. There are some lower expertise builds that we do that we can almost totally automate with agentic workflows and AI. And that touches, if you were to see it, touches six layers deep of the AWS tech stack from Bedrock to AgentCore to DynamoDB, et cetera. And I think that's really going to change the game for these cloud providers, these hyperscalers, where you're going to see that usage all the way through go up. You mentioned earlier data has gravity, and the amount of data that these agentic workflows are going to start to create and then have to start to consume within a company is going to be a game changer.
John Furrier
>> Trevor, final word. What are you optimizing for? You guys looking to hire, get some funding, you got ... Get the plug.
Trevor Hansen
>> Yeah. There's a couple of things. If you're a services partner where you're a consulting company, a GSI or a managed service provider and you're looking for a technical orchestration stack, reach out, we're excited to partner with you. Customers, as you're going through this journey, we are willing to jump on any conversation with no expectations. Give us the hard feedback, because we want to iterate on the product. From an investor perspective, as I mentioned, our goal is really to make sure we unlock long-term value for our investors, our partners. And the partnership with AWS has been phenomenal, so we're really excited about this journey.
John Furrier
>> What's your five-year proforma saying?
Trevor Hansen
>> Bring us back on five minutes to have the conversation.
John Furrier
>> Thanks for coming on, Trevor, Brandon. Thanks for Serge. I'm John Furrier, host of theCUBE here in our NYSE AI Factory series, of course, the NYSE Wired program at CUBE Original is an open network of innovators and leaders, making it happen and bringing in the future of generative AI and agentic. Thanks for watching.
>> Palo Alto Studio connecting Silicon Valley and Wall Street. I'm John Furrier, with Dave Vellante, my co-host. Welcome back to theCUBE here. I'm John Furrier, your host here at our New York Stock Exchange CUBE Studios, part of our East Coast hub, of course, connecting Silicon Valley from Palo Alto to Wall Street. It's our AI Factory series, part of our NYSE Wired program at CUBE Original. Got two great friends on the CUBE. Former Amazonians now starting a new venture. Again, coming out of stealth, kind of hitting the agent market. We got Trevor Hansen, founder and CEO. Blunom.ai. We'll get into the name in a second. Serge Shevchenko, who's the co-founder and head of revenue. The pressure's on you. Serge, great to see you.
Serge Shevchenko
>> Yeah, likewise.
John Furrier
>> Trevor, congratulations. This is your first public launch video here in theCUBE. Appreciate you giving us access. We've known each other from the Amazon day, so congratulations.
Trevor Hansen
>> Yeah, thanks, John. Firstly, it's been a tremendous journey. We've partnered multiple times throughout the journey, seeing customers go through the cloud migration now trying to figure out how to implement AI. I spent a lot of time deeply understanding the problem statements across companies trying to securely adopt this, and no better way to solve it to go build the product myself. So, being really passionate about building this sovereign AI strategy for customers. At a very high level, we realized that enterprise customers and every persona is sitting there trying to figure out how do they invest in this technology? How do they keep it safe? And then more importantly, how do they standardize? We came to realize that having a sovereign strategy is critical so that you can actually understand where you're deploying your stack, your datas, your own data. The model providers you're using, you have controls and boundaries around that. And that's what we set out to do is ensure we can equip customers with that technology.
John Furrier
>> Serge, talk about the unique Amazon position, because Amazon sees everything. You mentioned sovereignty. It's the hottest areas where the most of the infrastructure build out is where the cloud intersects. I mean, cloud sovereign has been around for a while, but now AI sovereignty's coming in. It speaks that you guys had a good visibility. What did that enable you to do, and how have your combined experience collectively bring you to this opportunity?
Serge Shevchenko
>> Yeah, I mean, I think spending time with partners and customers over about a decade. So, collectively we've been doing this for about 20 years. We noticed that the AI bottleneck was not just services or security or cost management. You're seeing some of the articles about organizations facing these challenges. That the challenge is actually the lack of the combination of both, proper AI orchestration and services, to really approach this new software development model like we talked about.
John Furrier
>> Trevor, you've seen a lot of the corp dev, product work, partnerships. Amazon Web Services probably was probably the best success story that I've seen in my lifetime. Now we've got the whole AI thing playing out too. How it's gone from just such a humble beginning, misunderstood as Andy Jassy would say, to this massive ecosystem, major enterprise penetration, just the growth. And okay, that hits it. Successes are there. Then you got this whole, that's the substrate now we're building on. Now you've got the agents coming over the top that's now globally distributed. What is unique about this venture? What jumped out at you? What made you guys start the company?
Trevor Hansen
>> Yeah. I mean, the first is partners play a critical role in this journey. So, there's existing technology providers, ISVs and consulting companies that are helping these customers migrate, modernize, unlock that value. The problem is when you're implementing this technology in isolation, there's a couple of problems that come up in the enterprise. You have shadow IT popping up in every line of business, because they're all procuring AI software individually, and with good intent.
John Furrier
>> Channel AI or IT?
Trevor Hansen
>> Well, both.
John Furrier
>> Both.
Trevor Hansen
>> It's both, because this technology-
John Furrier
>> IT has become the cloud, basically.
Trevor Hansen
>> Yeah. This technology isn't isolated. It actually integrates with your data and you need to make sure that you've got the right security protocols in place so that the agent doesn't index your whole database and spin up thousands of dollars of unnecessary cost, let alone pulling IP into a model provider where you want to expose your company. So, ensuring that the company has the right control plane in place and they have the controls in house allows them to be effective at evolving the business from a traditional architecture to be agentic. And when we started building this, we realized that partnering early for success was critical as well. So, working with consulting companies that give them the ability to choose their model provider, choose their cloud and choose the outcomes they're trying to invest in.
Serge Shevchenko
>> Yeah. And you talk about AI factories often here, and I think we're also being mindful about the infrastructure and the hardware layer as well, ensuring that agents are building on the right hardware and the most efficient hardware as possible.
John Furrier
>> On the AI factories, how do you plug into that? Because one of the pain points you were just mentioning, Trevor, is that the enterprise adoption is chaotic and it's total chaos, not even managed chaos. I mean, cloud, shadow IT was managed chaos, because essentially I was basically going around bottleneck blockers and organizations and getting a better deal in the cloud. That was good chaos, but it was more of a IT disruption. Shadow AI is just totally chaotic. Things can run wild, grab credentials, and the databases are involved. I mean, it's essentially a data sprawl, uncontrollable.
Trevor Hansen
>> Yeah. And the bigger risk for leaders at companies is they actually don't even have visibility into this. What agents are accessing, what data at what cost in their business. And when they sitting there saying, where are we deploying AI technology today? I'm pretty sure most CIOs and COOs are not able to actually show you a clean list. The risk starts all the way in the existing data centers where they're busy deploying technology to harness existing data. What we've done is actually looked at every layer of the stack and we partnering with infrastructure providers so that we can do cost optimized deployments within a customer's data center, so that when they're going through migration and modernization, they have a sovereign AI control plane that they can migrate and over time harness powers in the data center and in the cloud.
John Furrier
>> Go ahead.
Serge Shevchenko
>> Enterprises want the ability to fully control AI, not just in the cloud, but in the data center as well. And we give them the ability to do that, and the portability that when they are ready to go to the cloud, they can do that.
John Furrier
>> So Blunom, B-L-U-N-O-M.ai is the name of the company, Blunom. Serge, you mentioned this kind of like a genome. My son just did a genome sequencing and you see all the DNA and genetics. As mutations happen in the DNA going back to the origination. Why the name? Explain the name Blunom.
Serge Shevchenko
>> That's a great question. I don't know if you saw the new Fed chair, Kevin Warsh, just talked about how AI will just be called business. And we believe that through our platform, as we provide every line of business leader the ability to build, manage and deploy AI safely, securely, manage costs, we are truly like genome to the body, we are going to bring businesses to life.
John Furrier
>> And there's a lot of DNA in there.
Serge Shevchenko
>> Yeah, and to have-
John Furrier
>> That's an Oracle database from them, but system of record. Don't touch it. It's all our Salesforce stuff. So all these databases, this is why SaaS apocalypse is totally overblown, because that data gravity and mode, you can abstract that away. Now, they got to build better products. It's like the iPhone versus the iPod. They're two different things.
Trevor Hansen
>> Yeah. And to add to the Blunom, so blue being a color you trust from traditional IT, but the nom is the genomic aspect of your business. You need an antivirus system that the nervous system that's connected. One thing that we're really excited about is having a cognitive layer that allows the business to centralize knowledge and learnings. You talk about SaaS. We have over 150 different ISVs integrated into the platform that the agents natively understand how to interact with. When you're building an orchestration layer asking a simple question of your AI agents, that agent is able to actually go interact with securely with different software companies where you already have business users operating today. The valuable part is that you're able to deploy this technology to both technical and business users. We see a lot of companies today, they're building specifically for the technical persona and the developers. We believe that equipping every line of business, particularly the business users with a piece of experience in the AI that can drastically improve their productivity is going to be critical, but you have to do it with the right controls in place.
Serge Shevchenko
>> That's right. Well, and also just to touch on that, giving the CISO the ability to manage these agents, regardless of which line of business they're being deployed in. That's some of the challenges our customers have been telling us. They're seeing a lot of these agents being managed, built, deployed in different lines of business, and they have no idea how to manage not just cost but intra-agent communication or department communication east to west, which we're solving for today.
John Furrier
>> It's an integrated approach. The C-suites involve the line of business you mentioned. We're seeing that in the data from theCUBE and theCUBE research. It's not just an IT transformation, it's a business model transformation, it's a complete thing. You mentioned the partnerships. Knowing you guys were in stealth for a while, I know you were very kind of low key and I didn't even know what you were working on until I just found out today, so thanks for sharing. But how much work has gone in? Take us through from origination, how much momentum have you did? What work have you done integrating 150 ISVs?
Serge Shevchenko
>> Yep.
John Furrier
>> That's pretty significant. We're going to have Brandon Kissinger on, who's one of your first customers. You've done some work.
Serge Shevchenko
>> And partners.
John Furrier
>> Yeah. So take us through, because you got a lot of traction there. You kind of built it out. Take us through that.
Trevor Hansen
>> Yeah. I mean, what I saw firsthand is no enterprise customer is able to solve this problem alone. And they're bringing in specialized technical individuals. They're bringing in consulting firms they trust, so multi-partner collaboration to unlock value is going to be critical.
John Furrier
>> You're building an ecosystem stack basically into your layer?
Trevor Hansen
>> And that's something we're proud about is that we ecosystem orchestrators as well is that the channel from the beginning was a priority at the top of the board for us is realizing that we should not just go in alone, but partnering with software companies that deeply understand AI cloud migration modernization.
John Furrier
>> All right, so here's a question for you. In the training to inference evolution, everyone was one point into training, buy a bunch of GPUs, train the heck out of the data and then comes interested away. I bought that rack and I got to redo it. We're seeing a similar thing play out for agents. I think I nailed my layer and then all of a sudden, wait a minute, I need harmonization. I need compliance. Are you seeing the same thing play out? And what should customers do? Because if I could swap out models, I need to create some fusion.
Trevor Hansen
>> Yeah. I mean, now a couple of years into adopting AI and agents, you now realize that the infrastructure layer is going to be the most important aspect of this deployment is ensuring that this cross-functional collaboration across your business, that the models can communicate with each other. And more importantly, you're not creating data silos or AI silos in your business. So, having a platform that can connect the whole business and every agent can speak South to West across the business.
John Furrier
>> And Serge, talk about the model piece, because what you're seeing is small language runs, which we called four years ago, so we got that right. But now we're seeing real time swapping out models because of either economics or capabilities. It's like that scene in The Matrix. Upload how to fly a helicopter. You might need a model today for this.
Serge Shevchenko
>> Yeah. If agents are airplanes, and we consider ourselves as the FAA, right? We're making sure that people are not shipping a box from one city to another or perhaps even one neighborhood to another using a 737. You don't need to do that, swap it to a local model. Or perhaps train your own model and have the ability to sort of swap models based on the user need.
John Furrier
>> So, you're building intelligence into the layer.
Serge Shevchenko
>> Very much so.
Trevor Hansen
>> And to put a pin on that is, the more the platform gets consumed within their own environments, consumption makes recommendations. So, it actually starts recommending that the model you're using is overkill for email summarization and you could actually improve your cost by 70% in token usage, and it makes the recommendations to the business to swap out a model.
Serge Shevchenko
>> And add another layer to that, because model modularity is important, but add another layer, which is model modularity and hardware, suddenly you're looking at significant cost savings on the customer side.
John Furrier
>> All right, guys. So talk about the status of the venture, financing, headcount, what's your goals you guys hiring? What's the plans? Give a little stats on inside the numbers, what's going on or what's going on with the company.
Trevor Hansen
>> Yeah. I mean, we first set out to make sure we deeply understand the problem statement, build the right solutions, the right partnerships and engage in true business outcomes selling. Our goal is not to just go when drop in a product and step back. We really want to make sure we have success at the front of our partnerships and our customers. So, our first call was to a partner, and we're really proud that we built that out first. We're actively in conversations with investors and we're excited about those, but we're in the long run, we want to make sure we create compounding value.
John Furrier
>> So you have not done a big VC round yet?
Trevor Hansen
>> We're actively in conversations with venture capital and investors, and it's going to be a key part of our growth story, but customers and partners will always be our first points of entry.
Serge Shevchenko
>> Correct.
John Furrier
>> Great guys. Congratulations. Trevor, we just had a great session with Serge. Now Brandon Kissinger, the executive chairman of SnapSoft, your early design partner, and I love that you're on, Brandon. Thanks for coming on theCUBE and participating. A startup success is make or break by their entry strategy, right? You win or lose by your risk management, but it's a gut feel. So, smart money always gets a nice partner, design partner they call it. That's what you guys were. Explain the relationship that you guys have and how'd that come together?
Brandon Kissinger
>> Yeah, absolutely. And John, thanks for having me. I ended up sitting down with Serge over a cup of coffee. I told me about an idea him and a colleague had, started as going to kick off and we deploy agentic agents in the AWS ecosystem all the time. We're a premier tier AWS partner, been around since 2016. And what's always fascinating to me is you got to take a stance on this and truly believe that when you deploy agentic agents, there has to be a control plane. There has to be governance. You have to think about token economics if this thing is going to scale. And I was blown away when I sat down with these two and got to see the platform firsthand, and we've adopted it internally and for all of our customers.
John Furrier
>> And Trevor, you were detailing out with Serge about the vision and we talked about you guys pedigree at AWS. Now you have an AWS partner tier one. When you look at that, the cloud game was hard. If you look at like the growth of cloud, it was over a decade and a half or more than how you look at it. But now the AI thing's shooting up even faster, built on cloud native. I guess my question for you guys is, as you see the market surge up with the AI, you still got to run in cloud, but the on-prem piece has token economics built in there, because all AI teams will get built on premises first, almost like a land environment given the cost of tokens. So, you're starting to see actual enterprise workers get enabled for the first time using the tools as a utility and they're burning through their token budgets. So you're going to see a ton of on prem where the data is, but the cloud's not getting any smaller either. And now you've got international with sovereign cloud. So now there's a whole nother architectural reset. So the question I have for you guys, how do you see that architectural reset? Because it's not, that throw away the old, it's building on top of. How do you guys fit into that narrative?
Trevor Hansen
>> I mean, I can add to start. So I believe meeting a customer where they are in their journey of from their data center to the cloud. So they've invested significantly in existing infrastructure applications, data that they house internally that you don't have to wait to unlock that value in the cloud out the gates. What we have built as an orchestration platform that can drop into the data center with our infrastructure partners. And the key part there is partnering with services partners that can help you understand your current estates in your data center and help you be able to surgically identify workloads to move to scale into the cloud. And that's where the Better Together partnership comes in is having a technology platform that is sovereign and a services company that you trust that can help you through this transformation.
John Furrier
>> Brandon, you've seen a lot of the lift and shift, now the microservices come out, higher level services from Amazon. As the AI comes in, how's that intersecting from your standpoint?
Brandon Kissinger
>> I'll tell you the free cash flow that it opens up for a business is where I see so much of the importance when it comes to AI, and especially Agentic AI and its workflows, especially when you can start to operationalize it at scale, which is again what the Blunom platform helps us to do. But whether the customer's in the data center, whether they're in the cloud, I see this huge opportunity to migrate payroll costs into cloud consumption, usually at a discount 60, 70% of what you would have spent otherwise. And it's fascinating to see the level of human productivity that goes up in the company. And I'm sure like every business owner, I got folks whose sole job is to put revenue money at the top of the funnel, and then I have whole teams of people who need to bring it down to the bottom. And the more of it that gets at the bottom, the more I get to reinvest in growing the company. And that's the story I talk to with customers for why they adopt and go down this agentic journey.
John Furrier
>> Expand on that, because this is where the revenue action's coming in. If you look at all the agent successes I just did interview this morning with a founder of a company that's been around during even the big data days, and he's in the governance area, he's crushing it. But the revenue is really, for the ones that do agents right, see revenue, not just cost takeout. And now you bring up the other issue of, where's the budget going to come from to buy more tokens or build more teams? So, the economics become big and I really don't like the word tokenomics because that's just cost of tokens. There's other costs in the business model. Could you share your thoughts on how you see that playing out, because you were hitting on some of those things where, "Okay, yeah, I got, but I still got to reinvest."
Brandon Kissinger
>> Spot on, spot on. And I have feelings about it, And it's one of those things, there's a bit of a Groundhog Day. I was at the AWS Partner Summit sitting around the room with a bunch of guys and gals just like me who run SI partners, some ISVs, and it was fascinating. They got asked the question, "What would it take for you to implement more agentic AI in your customers and in your own company?" And without question, everybody came up well that we got to clean up the data first. And what's fascinating, because this is a change over the last few months, you don't have to clean up the data anymore. We're building agentic agents that actually complete a workflow. And when I say agentic agents, I'm not talking a singular agentic agent, I'm talking we're building full stack agentic applications that finish a process end to end. And so those token economics become critical when you're looking at that actual activity that person is going to do. Sometimes accuracy has to be higher. Sometimes it can be lower. You can cost save here, you can cost save there. Sometimes you have to spend a little bit more. You have to make sure that if you build out an agentic workflow to do a creative type of project, like write an email or a paper that they're not submitting it 47 times to change one sentence.
John Furrier
>> I mean, the cloud game was great. Dave Vellante and I would always joke about pay by the drink, USC pricing. When you get to the agents, they're drinking a lot. So the token thing, I mean, all kidding aside, this is policy routing, this is service level agreement like thinking. This is like an OS, it's got scheduling involved. I mean, this is a complex system.
Trevor Hansen
>> To add to that quickly. I think as leaders at a business, you want to see P&L impact. Your orchestration layer shouldn't just be executing the workflows and keeping those agents secure, but more importantly, it should be surfacing up recommendations for efficiency. And that is purpose-built workflows, more deterministic, what model providers should be plugging into your agents and you should be seeing that efficiency gain. So as your token spikes, essentially you should be seeing it continuously optimized. On the top line, that's where services partners come in is directly partner with the CEO and say, "What are the key priorities for the business? Where should we be investing the efficiencies we've gained?"
John Furrier
>> I think you guys are hitting an organizational kind of like a Venn diagram, everything's coming together, the C-suite's changing their roles, the engineering teams, partnerships. And then at the middle is the AI builder. I call it the AI builder persona, because it's not just a developer, it could be a business person. In the C-suite, the line of business owner's elevated, and now the C-suite's got the CFO and HR involved, CSO and CIO are super positions.
Brandon Kissinger
>> Absolutely.
John Furrier
>> And then the platform engineering and the data science, they're all coming together. So, in the middle is the AI builder. If you believe that, you say, "Okay, that's going to be the killer persona." How do you see that operating model working, and what attracted you to Blunom? Because that's the optimization point, because if I'm a builder optimizing payroll or closing the books in the C-suite or running platform engineering, I'm doing, I'm the builder.
Brandon Kissinger
>> So putting in these agents all throughout your company is like hiring hundreds of interns and giving them access to all your critical systems. These agentic workflows are not George Foreman grills. You can't just set them and forget them. And that's what attracted us to the Blunom platform the second I took a look at it. And the orchestration layer and the visibility it provides to my engineers as we build, as well as our customers when we deploy it and manage those is phenomenal. And every business does this. Every year they sit around and they do forward planning for the next year and there's KPIs that all of the C-suite, those line of business owners decide to hit. And then they ask for all the budget and headcount to be able to hit those KPIs. We did the same thing internally at SnapSoft. And then I sat down with the team and figured out which of the future headcount could we replace with agentic workflows. And it was a fascinating process to be able to show the ROI that way.
John Furrier
>> How long was the list? Must have been a huge list. And then let's take these three.
Brandon Kissinger
>> If you think of Snap, we came up with 16 Agentic workflows across my LOBs just off the top of their head, and then we started to build it out. But there was a key thing we learned in that process is that somebody's job has to be to manage those agents and those workflows full-time. It could be internal at the company if you have it, or you can outsource it to a partner like us to be able to manage them, build them, identify and build them and manage them for you. I sort of joke the CFO and the COO are my two biggest supporters at the customers we talk to, because those are the ones who always have-
John Furrier
>> Well, they can see the shadow AI hit their department. That's why shadow IT really drove the cloud business because the guys knew, "Hey, I put my credit card down, I'll get the prototype up, then I'll prove it and then I'll get budget." Now that's happening in every department like, "Oh my God, this is actually real."
Trevor Hansen
>> I want to come back to your question, which is, the reason why all these personas are gravitating towards each other to say, "How do we solve this together?" To me as an indication that businesses are going through a big transformation. They're trying to figure out how as a leadership team, they're going to invest in change and securely adopt AI. So when we think about our platform and the way we built it, we thought very meticulously about the CEO's AI strategy, the CFO and how we give them financial control, the COO on how to create orchestration and essentially a workflow management platform that could deploy to every line of business user, not just technical, but empowering every business user to be efficient.
John Furrier
>> Okay. So Brandon and Trevor, talk about where you guys intersect and what that means for people watching. Are you in business today? How do people engage? What's the status of the go-to market? What are you looking to add from a tech stack perspective?
Trevor Hansen
>> Yeah. So June 9th we're going into public preview, so please reach out to us, Blunom, also our partner, SnapSoft. But I would leave every leader watching this with three key takeaways. The first is take back control of your AI strategy. You have to think about a sovereign AI strategy for the long term. And when I say sovereign, I'm saying, bring back some control in house so that you have the controls, the levers, and you can see a future with the right platform. The second is partner early for success. You have to bring in specialized skilled people that share the same outcome interests that you have that are able to help you meet you where you are and essentially transform. And then finally is empower and democratize every person across the business. Stop focusing just on developers in the IT department. Your business users in the line of business are all trying to find technology to help them be more productive. The problem is they're all buying off-the-shelf vertical solutions. They're creating shadow IT for you and also limiting your visibility. So standardize on a control plane that empowers you.
John Furrier
>> Yeah. I'll say get some managed chaos, not full chaos. Brandon, talk about the execution of all this, because at the end of the day you got to execute. I mean, I can buy that strategy all day long. Sign me up. I want to take that hill, put AI and everything in my business. But to get to a system of execution, let's talk about sovereignty is hard. Talk about what you guys are doing and how this fits in as you guys execute, because you guys are also transitioning in real time too.
Brandon Kissinger
>> Yeah. John, that's the right question to ask. We are an all-in AWS partner and they have phenomenal programs to be able to make this easy to implement for customers. If you think about the three-legged stool for why somebody does or doesn't do something, expertise in house, the budget, and then of course the bandwidth. Not everybody has employees sitting around not doing anything ready to experiment. I'll tell you the way we've done it and the way we've packaged it up is if you followed me down, my KPI to budget and headcount conversation, we go into a customer, we talk to the different lines of businesses. We look to identify one agentic workflow end to end with the highest rate of return for the company, but the easiest to implement. And this is going to involve us building multiple agents with multiple integrations. Blunom does a phenomenal job with the integrations. And then we build that Agentic workflow in 30 to 45 days soup to nuts and it's ready for UAT, user accepted testing. We give it to the teams to play with it and give us feedback. Then we run you all the way back and then I say, "John, we did one for you. The team is happy. They're using it. Give me, let's do the next four to six." And we're going to run it through over the next two.
John Furrier
>> And you guys are intentional at the targeting of that workload.
Brandon Kissinger
>> Yes.
John Furrier
>> This isn't a risk management side project. This is let's target where revenue business value is hitting.
Brandon Kissinger
>> And we need to figure out, we need to decide on what that ROI is at the very beginning. I'll tell you internally, we have our AP and AR process automated with agentic workflows, our allocation process with agentic workflows. So many of our business unit and teams enabled with agentic workflows and the ROI becomes clear. But get your first one, I would tell everybody, get your first one done and then we'll build out a half a dozen for you.
John Furrier
>> Do you see the kind of makeup of an individual ... Let me rephrase. Is there a threshold for implementing that kind of full stack? When does someone qualify to and leverage a platform, or how does someone become AI-native with it? What's the entry point? I guess that's my question. How would you guys talk about it?
Brandon Kissinger
>> Trevor might have an opinion on this one, and it depends on the workflow. For the one I'm talking about specifically, SnapSoft is a few hundred employees. I have two dozen agentic workflows that I can deploy, and it's the ones who ... The companies that do get enabled by AI are the ones who are going to have more free cash.
John Furrier
>> It doesn't have to be a big enterprise.
Brandon Kissinger
>> It doesn't have to be. I think even a lot of the .
John Furrier
>> SMB.
Brandon Kissinger
>> Absolutely.
John Furrier
>> Cloud-native company, obviously with Amazon.
Trevor Hansen
>> I think to add is what's nice about our platform is you don't have to have enterprise capability deployed out the box for an SMB. SMBs don't have a technical or an AI department.
John Furrier
>> So, we have DevOps at theCUBE. We have full AWS stack. For us, we could leverage it.
Trevor Hansen
>> Yeah, exactly.
John Furrier
>> We have workflows, not a lot.
Trevor Hansen
>> Yeah. So, when you look at customers that are trying to implement this, digital native are going to be moving a lot faster. You have 500 to 2,000 users. They're the ones leaning and investing in this technology. Ideally, they've been building some version of a stack in house, and they realize they better serve moving those engineers to customer-facing experiences versus trying to build their own orchestration plan. So I would recommend, if you are sitting there thinking about your economics and the amounts of developers that you're building in house, what I see is, it's really compelling when your developers moving at 10 times the speed in the past, but the thing is, what are they working on as an outcome? Are they actually building a product that's going to impact your P&L? If they're not, bring in Blunom, bring in SnapSoft. We'll dive in and help you identify use cases.
John Furrier
>> Use I think the P&L impact's huge. I guess my final question for you guys, because you've got great visibility, I'm curious. There's a huge set of customers out there that are on AWS and have had all the goodness of the cloud doing great. I mean, they're building platforms on top of AWS. Now they have to architect the journey. What's your advice for people? Because I'm imagining with all the inside baseball you guys have of AWS, there's a lot of nice intersections between the on prem sovereignty global footprint, dealing with regions, dealing with fail back over, data privacy, revenue generation in country. What's your advice for customers out there? Amazon Web Services customers and other clouds.
Trevor Hansen
>> To add, and then I'll hand to Brandon, you need to lean in with your AWS account management team and tell them your long-term strategy and call out partnerships that are critical to you. And that's something we've really enjoyed is that the account teams are partner friendly. They are co-selling to help unlock cloud value, and there's programs that help with migration, modernization, AI adoption, AI assessments, et cetera. So really making sure you're leveraging the channel of AWS. So for our engagements for every customer, we're going to contact your AWS account manager, say, "Hey, this is us. We're partner obsessed. We're coming to help you unlock your cloud journey with this customer, and we're going to be in the accounts for the long term."
John Furrier
>> I guess my question is, where do I plug in to AWS? I got my partner management, but I'm thinking of architect. Do I just plug into the stack?
Trevor Hansen
>> And that's where our business outcome orchestration part of our businesses is we immediately dial and connect up all the different stakeholders together. So, if you don't know the right place to go, contact us and we'll help you-
John Furrier
>> Can I do self-service?
Trevor Hansen
>> Yes. There's a self-service you can authenticate with Microsoft or Google on the platform and your users can populate, but at the end of the day, you really want to make sure that you're transforming with partners for success. So as much as you can make your own progress, which the platform provides and a SaaS-like experience as well.
John Furrier
>> So will you guys be in the marketplace?
Trevor Hansen
>> Yes. AWS marketplace is going to be a key point here is really helping those enterprise customers draw down their investments in-
John Furrier
>> So, I could consume through the marketplace?
Trevor Hansen
>> Correct. So marketplace private pricing agreements or just standard subscription is going to be a key part, and we're excited about-
John Furrier
>> And how are you guys thinking about revenue model just while I got you here?
Trevor Hansen
>> So revenue, and that's why we're moving away from per seat licensing. We believe we're going to be pretty disruptive in the economics of SaaS and software. We looking at platform and outcome based pricing. So really making sure that we are pricing our model for the long term and seeing customers invest more in us because they see results.
Brandon Kissinger
>> The more customers I onboard to it, the more they charge me.
John Furrier
>> So Brandon, I got to ask you a question since you're here, because I'm fascinated by the ecosystem evolution. How do you see the cloud ecosystem evolving to this AI-native ecosystem? Because the game is still the same, but the game has kind of changed. APIs are great, but now you've got agent connections, you got a lot more co-design going on, a lot more engineering in the partnerships. How do you see the success formula in the ecosystem to be successful building on top of cloud and to take advantage of the AI native?
Brandon Kissinger
>> Oh, well, I'll talk about it maybe from a personal standpoint as an SI partner. And I think this is where a lot of companies have to wake up and get real that there's going to be parts of your business and part of the things you do that you're going to have to cannibalize, or somebody's going to eat your lunch. And we looked at it internally even for us. There are some lower expertise builds that we do that we can almost totally automate with agentic workflows and AI. And that touches, if you were to see it, touches six layers deep of the AWS tech stack from Bedrock to AgentCore to DynamoDB, et cetera. And I think that's really going to change the game for these cloud providers, these hyperscalers, where you're going to see that usage all the way through go up. You mentioned earlier data has gravity, and the amount of data that these agentic workflows are going to start to create and then have to start to consume within a company is going to be a game changer.
John Furrier
>> Trevor, final word. What are you optimizing for? You guys looking to hire, get some funding, you got ... Get the plug.
Trevor Hansen
>> Yeah. There's a couple of things. If you're a services partner where you're a consulting company, a GSI or a managed service provider and you're looking for a technical orchestration stack, reach out, we're excited to partner with you. Customers, as you're going through this journey, we are willing to jump on any conversation with no expectations. Give us the hard feedback, because we want to iterate on the product. From an investor perspective, as I mentioned, our goal is really to make sure we unlock long-term value for our investors, our partners. And the partnership with AWS has been phenomenal, so we're really excited about this journey.
John Furrier
>> What's your five-year proforma saying?
Trevor Hansen
>> Bring us back on five minutes to have the conversation.
John Furrier
>> Thanks for coming on, Trevor, Brandon. Thanks for Serge. I'm John Furrier, host of theCUBE here in our NYSE AI Factory series, of course, the NYSE Wired program at CUBE Original is an open network of innovators and leaders, making it happen and bringing in the future of generative AI and agentic. Thanks for watching.
John Furrier