This live conversation at Boomi World 2026 features Eddy Montalvo of Boomi, vice president North America alliances and channels; Nikhil Gauba of Cognizant, senior director client artificial intelligence, AI and application programming interface, API strategy; and Yogeshwar Pandey of Cognizant, chief architect and digital practice leader. theCUBE Research hosts John Furrier and Gemma Allen moderate the discussion.
The conversation addresses enterprise strategies for agentic systems, integration, API strategy, agentic architectures, control plane design, connectivity, data activation and pilot management. Gauba emphasizes the need for new operating models and governance as agentic systems scale; they state that enterprises must define human-in-the-loop roles and establish guardrails for agent decisions. Montalvo underscores Boomi's role as connective tissue and a control-plane enabler; they highlight the platform's capacity to simplify integration and accelerate data activation. Pandey focuses on practical architecture considerations across industries such as banking; they advise disciplined pilots, FinOps for AI cost management and measurable return on investment as critical factors for successful enterprise adoption and digital transformation.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
Boomi World 2026. If you don’t think you received an email check your
spam folder.
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 Boomi World 2026
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 Boomi World 2026.
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
Boomi World 2026. If you don’t think you received an email check your
spam folder.
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 Boomi World 2026
Please sign in with LinkedIn to continue to Boomi World 2026. Signing in with LinkedIn ensures a professional environment.
This live conversation at Boomi World 2026 features Eddy Montalvo of Boomi, vice president North America alliances and channels; Nikhil Gauba of Cognizant, senior director client artificial intelligence, AI and application programming interface, API strategy; and Yogeshwar Pandey of Cognizant, chief architect and digital practice leader. theCUBE Research hosts John Furrier and Gemma Allen moderate the discussion.
The conversation addresses enterprise strategies for agentic systems, integration, API strategy, agentic architectures, control plane design, connectivity, data activation and pilot management. Gauba emphasizes the need for new operating models and governance as agentic systems scale; they state that enterprises must define human-in-the-loop roles and establish guardrails for agent decisions. Montalvo underscores Boomi's role as connective tissue and a control-plane enabler; they highlight the platform's capacity to simplify integration and accelerate data activation. Pandey focuses on practical architecture considerations across industries such as banking; they advise disciplined pilots, FinOps for AI cost management and measurable return on investment as critical factors for successful enterprise adoption and digital transformation.
Chief Architect - Digital Practice LeaderCognizant
search
John Furrier
>> Welcome back over to theCUBE live here in Chicago. Boomi World 2026. I'm John Furrier, host of theCUBE here with Gemma Allen as well, getting all the action, and the top story is how gen AI and now agentic enterprises will be changing their business looking for the right outcomes. We've got a great panel here, a hybrid interview panel with theCUBE. We love doing with leaders doing it, Cognizant and Boomi. We've got Nikhil, he's Senior Director of Client AI and API Strategy. Good to see you. Yogeshwar-
Nikhil Gauba
>> Thank you....
John Furrier
>> Chief Architect, Digital Practice Leader at Cognizant. You guys doing a lot of change management, a lot of modernization. Of course, we got Eddy here, Vice President of North America Alliances and Channels at Boomi. We got the Boomi, Cognizant dialogue here. Guys, the number one conversation we're having on theCUBE here at our NYSE studio and in Palo Alto and all around the world and in events is business transformation is legit, but it's hitting areas that it didn't traditionally hit. It's the business model. It's the deep tech impact and the developer angles. It's all three areas are popping at once. This is a huge deal because it's harder, but the tooling's getting better. We'll go to you guys first, as you guys look at these transformation, yeah, change management, people, process technology, we've heard that in IT. What's changed the most as it moves from an IT to a much more broader aperture?
Nikhil Gauba
>> Yeah. I mean, with the agentic ecosystems coming in, it's no longer a IT shift or IT pivot. I would say it's more of a business shift with companies and enterprise looking at how they can improve their business outcomes and ROI. So it's changing the business model and changing how we operate. So there has to be a shift in the operating model and the business model also.
John Furrier
>> Architecturally, what part of the playbook remains the same from the old way, and what's changed on the AI side architecturally, practice? What do you think?
Nikhil Gauba
>> I think the way we have been seeing this transformation where we build the enterprise integration and then those capabilities we exposed via an API. I think the next thing is more about the agentic coming and then leveraging those APIs are going to be a key and that's what we are seeing in our customer science.
John Furrier
>> Eddy, your CEO was up on stage. He went six minutes without saying the word AI, which I know is an internal joke. There's probably an over under inside Boomi and how long he can go, but he did that intentionally to build the suspense up to talk about the criticality of the moment we're in. Your partnership with these guys and the market's growing because you're touching more things.
Eddy Montalvo
>> Yeah. No, I think what's really interesting in this space right now and just how well Boomi is positioned to help drive the change with partners like Cognizant. I think the really exciting part is customers know about us for our connective tissue, how we've been able to help them connect applications across their stack. But now, to Yogesh's point, we went from the IPaaS, the APIM era, and now we're getting into really the agentic connectivity tissue, which Boomi has really made a strong push into the market with taking those connectors that we helped start the backbone of Boomi, but now making those connectors accessible via MCP.
John Furrier
>> The relationship with Cognizant, how's that going? What's that look like? What's changed with Boomi and Cognizant?
Eddy Montalvo
>> I think to something we were talking about earlier, I think you said it, that modernization's still the same, but the game's changing a little bit. And I think doing that with partners like Cognizant, we're working with a lot of customers, for example, in the banking industry and to the comment around modernization. Boomi is now in the forefront of a banking modernization project. We're no longer in the plumbing, so to say, and the connective pipes. They're thinking, how do we modernize the way that we're doing banking today? And Boomi is actually facilitating that across data activation, connective tissue with our connecting products, and then obviously again, on the agentic journey.
Nikhil Gauba
>> Yeah, exactly. I mean, we are talking to the customers on how they can improve their revenue channels for banking for embedded finance, so it's no longer connecting to their core systems. It's going beyond into more business outcomes.
John Furrier
>> Yeah, revenue.
Nikhil Gauba
>> Yeah, revenue.
John Furrier
>> And the AI generation, at least one thing that to me, on my observation, looking at all these events and talking to many people, it's a revenue conversation.
Nikhil Gauba
>> Exactly.
John Furrier
>> Yeah, there's cost takeout.
Nikhil Gauba
>> Yeah. Oh, yeah.
John Furrier
>> But there's more revenue upside, so that's why it gets the CFO's attention before.
Nikhil Gauba
>> Exactly.
John Furrier
>> So here's your budget, go do the IT project. Now, they want to lean in because shadow AI is hitting every domain. CFO, they're closing the books faster. They got to go to the system of record to execute. You got every department, sales, marketing, you name it, it's all happening. This comes back down to what you mentioned earlier about the operating model. I guess I have to ask, what is the operating system and what is the operating model on top?
Nikhil Gauba
>> Yeah. I mean, when I talk about operating model change, it's all about who takes the decision when agents take control. Is it human in the loop? Is it agent in the loop? How do we put the right governance and controls and insights on what decisions they are taking? There has to be a thought on how we shift to that operating model. We don't want agents to go rogue so there has to be guardrails and a referee. What Steve talked about, there has to be a referee who's governing all the rules, and that's where Boomi comes in and plays a vital role in how we set up the foundational capabilities.
John Furrier
>> Steve mentioned the SaaS-pocalypse. What he didn't mention, which I always talk about is that MIT study that came out last year that said most of the projects are failing, which I thought was a false narrative in my opinion. And I was talking to Dave Vellante in our podcast, I said, "David, AI is like giving your 16, everyone's a 16 year old driving for the first time with the parent's car. It's going to be statistically more fender benders than better drivers. And if AI is the car, technology is the car, the humans are driving. You're only as good as your one knowledge of the machine."
This seems to be a metaphor that's landing. What do you guys think about that? Because the change management piece is a human piece. How do you guys roll this out? Because you got to have the governance, the rules of the system of record. He mentioned, he said rules are ERP and CRA. He's not wrong. They could be abstracted away. Talk about this human role in that driving the innovation.
Nikhil Gauba
>> Yeah. I mean, many of customers might be thinking we can just replace humans with AI, but that might not be true so humans have to be in the loop. Every decision what agent takes, there has to be a controlled mechanism and visibility into what decision has been taken, so you cannot just replace humans with the agents. There has to be humans and agents working together in collaboration. Maybe we need more roles for humans, so there might be AI human role which is monitoring what all agents are doing and pressing the kill switch if it goes wrong. You know what happens if it starts leaking of financial information about the company or the customer information, and it might have happened we are seeing many cases now. Humans are important, they have to be in this ecosystem. It cannot be just agent.
John Furrier
>> I mean, you're basically the kill switch, basically it's cyber resilience.
Nikhil Gauba
>> Exactly.
John Furrier
>> That's what we're talking about here. Okay, so take that to the next level because this is the number one conversation. You mentioned vibe coding, which everyone's been talking about. And everyone knows you got to actually plug it into an operational system.
Nikhil Gauba
>> Yep.
John Furrier
>> Talk about how enterprises are thinking about scaling. Because when you talk about scale, you have to know the governance, you got to get the trust, you have to run time on the Boomi side. And they got the observability piece, evaluating and now they got the new acquisition with routing models. I mean, this is coming together. How does a company execute an agentic strategy? How do you guys advise and how do you put that together?
Nikhil Gauba
>> I mean, the first thing is most of the agentic programs or projects are in the pilots. They are in the experimentation stage, so it has to go beyond that. That is where Cognizant and Boomi are partnering with setting up the right foundation for all the capabilities you talked about, be it controls and governance, be it guardrails and insights, the observability. That's where we have the reference frameworks. We can set the right foundation. And then how do we scale at an enterprise level by more horizontally is what we are trying to do so it is not just one project at one business line, it is how do we roll out across enterprise with the right foundation? And that's where Boomi's capabilities come in with all what Steve talked about.
John Furrier
>> I mean, Boomi's got great chops on integration. We've covered that at depth. On the pilots, is there a playbook? No one wants isolated, siloed pilots. In fact, I was talking to a CFO and she told me that she could tell things are doomed from the start because people just sandbag and throw the pilots out there out on the fringes. It didn't work. So there's a lot of pilot management on the wrong side. What's the right way to manage pilots?
John Furrier
>> The way I feel we should go about this is define your agent, and with a very defined scope for that. And then you can leverage your Boomi control plane so that you know the rogue risk of the agentic behavior. You can able to control, and the control pane will give you all the guardrails, the policy which you can apply to the agents. You can have a monitoring and the observability around it and then apply the kill switch when it is going beyond. I think that learning, you can infuse it and scaling further to this.
John Furrier
>> I mean, basically risk management's built in.
Nikhil Gauba
>> Yeah.
Nikhil Gauba
>> Exactly.
Nikhil Gauba
>> I think adding to that, is it creating a value? Is it creating an ROI? So agents are failing because there is no measured KPIs on that, and then has it the right foundation? Those two are more important for me.
John Furrier
>> I wrote a post last night that said that Boomi's entering in the rise of the control plane in Chicago, little bit of teaser for the keynote. Read his hand blind a little bit there, but I interviewed him earlier. He didn't reveal anything. Steve didn't tell me anything, but I put it together. But I did make a comment. I want to get your reactions. I think you hit a great point there. Most organizations think chatbots, they came in from the chatbot side, and yeah, they're putting out some wrapper stuff and you get a prompt, you get in some sort of action or outcome experience. But the people that are designing their businesses, redesigning their businesses are looking at things much differently. Architecture, governance, and ROI, that's the new execution playbook. What's your thoughts on that?
Nikhil Gauba
>> Yeah, and the cost of operations because tokens are being involved, so there has to be a cost management part to it. So you're right, I mean, these are the basic things. Whenever we think of an agent in a business process, it has to be there.
John Furrier
>> All right, let's talk cost. FinOps is just now getting standardized in the cloud world, but we're beyond the cloud.
Nikhil Gauba
>> Exactly.
John Furrier
>> Remember, cloud had cost optimization, the whole FinOps category emerged from cloud. Now you got AI, it makes cloud look like it's settled. I mean, AI is probably a 10X more complex cost equation. Thoughts on that where you see the, I don't know what to call it. It should be called token economics.
Nikhil Gauba
>> AIOps.
John Furrier
>> FinOps for AI, what would that look like?
Nikhil Gauba
>> Yeah, I mean-
John Furrier
>> How should people frame at least their mindset?
Nikhil Gauba
>> Token management and cost management as a capability has to be at enterprise level, and then how do we monetize the agents based on that cost has to be in the enterprise. It is not just cost management, but even the monetization play for the business line. I mean, it is very foundational. Many enterprise are not looking at it now, but like the cloud, when they see the cost exponentially increasing, that has to be an important part.
John Furrier
>> It's funny you bring that up. I was talking to a chief human resource officer. As you said, there's already internal politics around who has more agents in their organization. And then there's also an effort to evaluate the performance of those agents, and then so there's a reorg mindset going on. That's a technical policy. How do you guys view that? Because I think you're right. We want to have agents under organizations, and hey, I got a ton of agents, some are doing work, idle, are not performing. Potentially, okay, you could have more agents in your department. Joking aside, but that's happening.
Nikhil Gauba
>> Exactly.
Nikhil Gauba
>> I think the way you can able to control it is making sure the agents are very contextual so that the number of tokens you're limiting. And that's why I think the prompt engineering is coming also as a handy part so that you can able to train your agent how those prompt will be structured. That's how you're controlling. I think FinOps, one of the key point is early detection of those wrong prompts, and then not hitting all the way to the token or the LLM.
Nikhil Gauba
>> Right. I mean, even the frontier models, which are large language models, they are expensive. We are also recommending small language models which are most cost-effective and enterprises can use in more use cases, which they don't need to use LLM.
John Furrier
>> And it saves on compute and GPU costs.
Nikhil Gauba
>> Exactly. Companies are moving towards using SLMs and managing costs for the tokens and how they do the prompt engineering to Yogesh point is there is a lot of enablement to be done and we can save a lot of token costs by how you tell the functionality should be done by an agent.
John Furrier
>> And you guys are basically talking about why I brought up architecture. We're in an architecture strategy game, not IT. I mean, ultimately, because it's a systems, it's not... It's a system effect.
Nikhil Gauba
>> Yeah. I mean, if you look at the agentic architecture, there are seven or eight layers. It's not just you have built an agent. There are so many layers which goes beyond that so all of those layers have to be set properly with the right guardrails. It includes how do you write prompts? What are the prompt injection in that? How do you control the data which is coming out of your assets which are connected by Boomi? So it's not just agents and doing something, but it has to be right across all the layers.
John Furrier
>> And you guys-
John Furrier
>> Also like to add what Steve mentioned in his keynote, the connect. If your connect is mature, then you can have the right context. Really, the agents can hit the right context to those APIs.
Nikhil Gauba
>> Activating your data.
John Furrier
>> Activating your data.
John Furrier
>> I mean, I do like the concept of connect everything, get the contextual data and understand and route accordingly. I love the new acquisition and then the control point. I mean, that's the three Cs right there.>> Yep.
Nikhil Gauba
>> Yep.
John Furrier
>> All right guys, final question, I guess I'll maybe do a round table on this one. Eddy, you're going to weigh in from the Boomi side, but on the Cognizant side, people want to know what this event means because it's an inflection point for Boomi because they got a great track record and integration. I mean, IPaaS segment people that know that, know that's a lot of work that's gone in and they certainly have the bar high on doing that. But now it's shifting to real time, everything's being integrated in real time. When agents are being redeployed or policy is being set, you got to have, they got to bring that expertise to the table. We saw a peek of the future on the keynote. What's the new Boomi look like? If someone asks you, what is the new Boomi opportunity? Knowing what you guys know, because it's a new Boomi. It's a bigger market TAM, bigger use cases, more horizontal use cases and all that integration skillset comes to the table. What's your thoughts on this?
John Furrier
>> I see today the launch was Boomi Connect. I'm yet to see that, but I think already the enterprises where we were helping earlier when we were dealing with them, it used to take weeks and months to really develop those, but then Boomi Enterprise already gives a lot, many connectors and a lot of Boomi AI components are making already the ease to make those things within a week or within a day.
John Furrier
>> So faster integrations-
John Furrier
>> And now within minutes, faster integration.
Nikhil Gauba
>> I think more than that, it's the personas and stakeholders we are targeting. It'll be totally different now. We'll be talking to CFOs, business stakeholders, maybe sales and marketing. We are influencing and creating agents for business outcomes, not just IT outcomes.
John Furrier
>> It's a business conversation you say.
Nikhil Gauba
>> It's a business conversation.
John Furrier
>> Boomi up the stack.
Nikhil Gauba
>> It's a larger game, more money for us. And go Boomi and Cognizant.
Nikhil Gauba
>> Go Boomi.
Eddy Montalvo
>> In a good way, right? I think ultimately, I think it's such an exciting time for Boomi because we're putting ourselves at the center of the transformation efforts that we can do for our clients. So when we talk about activating the data, right? It puts you at the forefront of every conversation with every business stakeholder and say, "How do you really want to go about this transformation?" And luckily for Boomi, its toolbox has put us in the front and center of that problem statement and how we really go solve that opportunity.
John Furrier
>> Yeah. I mean, my big takeaway is governance is not just compliance anymore, it's operations, execution, risk. So system of execution will be the agents.
Nikhil Gauba
>> And that's very unique positioning of Boomi. They have taken the AI governance side of the house, which is the key for making your agentic journey very successful.
John Furrier
>> Show me people that like to work on governance. Hey, who likes governance?
Eddy Montalvo
>> We talked about it a few times. Going to do it in a federated model, because we know they're going to use a lot of different tools, right?
Nikhil Gauba
>> And building trust.>> Building trust, yeah.
John Furrier
>> It's like developers, they don't want to do that, they don't want to debug anything, but they have to. But compliance, if done right, makes everything fly.
Nikhil Gauba
>> Right.>> Yep.
John Furrier
>> Guys, thanks for coming on in theCUBE. Really appreciate it. Looking forward to getting more insights down the road. Appreciate your time.
Eddy Montalvo
>> Yeah, likewise, thank you.
Nikhil Gauba
>> Thank you so much. Thank you.
John Furrier
>> I'm John Furrier here at theCUBE. Again, governance, these new system architecture, architecture, execution, return on investment. That's the new criteria, not just prompt to get a little action and some outcome. The results are going to come down to revenue and ultimately enterprise wide scale. Boomi's going after it here in Chicago. Thanks for watching.>> Thanks, bye-bye.
>> Welcome back over to theCUBE live here in Chicago. Boomi World 2026. I'm John Furrier, host of theCUBE here with Gemma Allen as well, getting all the action, and the top story is how gen AI and now agentic enterprises will be changing their business looking for the right outcomes. We've got a great panel here, a hybrid interview panel with theCUBE. We love doing with leaders doing it, Cognizant and Boomi. We've got Nikhil, he's Senior Director of Client AI and API Strategy. Good to see you. Yogeshwar-
Nikhil Gauba
>> Thank you....
John Furrier
>> Chief Architect, Digital Practice Leader at Cognizant. You guys doing a lot of change management, a lot of modernization. Of course, we got Eddy here, Vice President of North America Alliances and Channels at Boomi. We got the Boomi, Cognizant dialogue here. Guys, the number one conversation we're having on theCUBE here at our NYSE studio and in Palo Alto and all around the world and in events is business transformation is legit, but it's hitting areas that it didn't traditionally hit. It's the business model. It's the deep tech impact and the developer angles. It's all three areas are popping at once. This is a huge deal because it's harder, but the tooling's getting better. We'll go to you guys first, as you guys look at these transformation, yeah, change management, people, process technology, we've heard that in IT. What's changed the most as it moves from an IT to a much more broader aperture?
Nikhil Gauba
>> Yeah. I mean, with the agentic ecosystems coming in, it's no longer a IT shift or IT pivot. I would say it's more of a business shift with companies and enterprise looking at how they can improve their business outcomes and ROI. So it's changing the business model and changing how we operate. So there has to be a shift in the operating model and the business model also.
John Furrier
>> Architecturally, what part of the playbook remains the same from the old way, and what's changed on the AI side architecturally, practice? What do you think?
Nikhil Gauba
>> I think the way we have been seeing this transformation where we build the enterprise integration and then those capabilities we exposed via an API. I think the next thing is more about the agentic coming and then leveraging those APIs are going to be a key and that's what we are seeing in our customer science.
John Furrier
>> Eddy, your CEO was up on stage. He went six minutes without saying the word AI, which I know is an internal joke. There's probably an over under inside Boomi and how long he can go, but he did that intentionally to build the suspense up to talk about the criticality of the moment we're in. Your partnership with these guys and the market's growing because you're touching more things.
Eddy Montalvo
>> Yeah. No, I think what's really interesting in this space right now and just how well Boomi is positioned to help drive the change with partners like Cognizant. I think the really exciting part is customers know about us for our connective tissue, how we've been able to help them connect applications across their stack. But now, to Yogesh's point, we went from the IPaaS, the APIM era, and now we're getting into really the agentic connectivity tissue, which Boomi has really made a strong push into the market with taking those connectors that we helped start the backbone of Boomi, but now making those connectors accessible via MCP.
John Furrier
>> The relationship with Cognizant, how's that going? What's that look like? What's changed with Boomi and Cognizant?
Eddy Montalvo
>> I think to something we were talking about earlier, I think you said it, that modernization's still the same, but the game's changing a little bit. And I think doing that with partners like Cognizant, we're working with a lot of customers, for example, in the banking industry and to the comment around modernization. Boomi is now in the forefront of a banking modernization project. We're no longer in the plumbing, so to say, and the connective pipes. They're thinking, how do we modernize the way that we're doing banking today? And Boomi is actually facilitating that across data activation, connective tissue with our connecting products, and then obviously again, on the agentic journey.
Nikhil Gauba
>> Yeah, exactly. I mean, we are talking to the customers on how they can improve their revenue channels for banking for embedded finance, so it's no longer connecting to their core systems. It's going beyond into more business outcomes.
John Furrier
>> Yeah, revenue.
Nikhil Gauba
>> Yeah, revenue.
John Furrier
>> And the AI generation, at least one thing that to me, on my observation, looking at all these events and talking to many people, it's a revenue conversation.
Nikhil Gauba
>> Exactly.
John Furrier
>> Yeah, there's cost takeout.
Nikhil Gauba
>> Yeah. Oh, yeah.
John Furrier
>> But there's more revenue upside, so that's why it gets the CFO's attention before.
Nikhil Gauba
>> Exactly.
John Furrier
>> So here's your budget, go do the IT project. Now, they want to lean in because shadow AI is hitting every domain. CFO, they're closing the books faster. They got to go to the system of record to execute. You got every department, sales, marketing, you name it, it's all happening. This comes back down to what you mentioned earlier about the operating model. I guess I have to ask, what is the operating system and what is the operating model on top?
Nikhil Gauba
>> Yeah. I mean, when I talk about operating model change, it's all about who takes the decision when agents take control. Is it human in the loop? Is it agent in the loop? How do we put the right governance and controls and insights on what decisions they are taking? There has to be a thought on how we shift to that operating model. We don't want agents to go rogue so there has to be guardrails and a referee. What Steve talked about, there has to be a referee who's governing all the rules, and that's where Boomi comes in and plays a vital role in how we set up the foundational capabilities.
John Furrier
>> Steve mentioned the SaaS-pocalypse. What he didn't mention, which I always talk about is that MIT study that came out last year that said most of the projects are failing, which I thought was a false narrative in my opinion. And I was talking to Dave Vellante in our podcast, I said, "David, AI is like giving your 16, everyone's a 16 year old driving for the first time with the parent's car. It's going to be statistically more fender benders than better drivers. And if AI is the car, technology is the car, the humans are driving. You're only as good as your one knowledge of the machine."
This seems to be a metaphor that's landing. What do you guys think about that? Because the change management piece is a human piece. How do you guys roll this out? Because you got to have the governance, the rules of the system of record. He mentioned, he said rules are ERP and CRA. He's not wrong. They could be abstracted away. Talk about this human role in that driving the innovation.
Nikhil Gauba
>> Yeah. I mean, many of customers might be thinking we can just replace humans with AI, but that might not be true so humans have to be in the loop. Every decision what agent takes, there has to be a controlled mechanism and visibility into what decision has been taken, so you cannot just replace humans with the agents. There has to be humans and agents working together in collaboration. Maybe we need more roles for humans, so there might be AI human role which is monitoring what all agents are doing and pressing the kill switch if it goes wrong. You know what happens if it starts leaking of financial information about the company or the customer information, and it might have happened we are seeing many cases now. Humans are important, they have to be in this ecosystem. It cannot be just agent.
John Furrier
>> I mean, you're basically the kill switch, basically it's cyber resilience.
Nikhil Gauba
>> Exactly.
John Furrier
>> That's what we're talking about here. Okay, so take that to the next level because this is the number one conversation. You mentioned vibe coding, which everyone's been talking about. And everyone knows you got to actually plug it into an operational system.
Nikhil Gauba
>> Yep.
John Furrier
>> Talk about how enterprises are thinking about scaling. Because when you talk about scale, you have to know the governance, you got to get the trust, you have to run time on the Boomi side. And they got the observability piece, evaluating and now they got the new acquisition with routing models. I mean, this is coming together. How does a company execute an agentic strategy? How do you guys advise and how do you put that together?
Nikhil Gauba
>> I mean, the first thing is most of the agentic programs or projects are in the pilots. They are in the experimentation stage, so it has to go beyond that. That is where Cognizant and Boomi are partnering with setting up the right foundation for all the capabilities you talked about, be it controls and governance, be it guardrails and insights, the observability. That's where we have the reference frameworks. We can set the right foundation. And then how do we scale at an enterprise level by more horizontally is what we are trying to do so it is not just one project at one business line, it is how do we roll out across enterprise with the right foundation? And that's where Boomi's capabilities come in with all what Steve talked about.
John Furrier
>> I mean, Boomi's got great chops on integration. We've covered that at depth. On the pilots, is there a playbook? No one wants isolated, siloed pilots. In fact, I was talking to a CFO and she told me that she could tell things are doomed from the start because people just sandbag and throw the pilots out there out on the fringes. It didn't work. So there's a lot of pilot management on the wrong side. What's the right way to manage pilots?
John Furrier
>> The way I feel we should go about this is define your agent, and with a very defined scope for that. And then you can leverage your Boomi control plane so that you know the rogue risk of the agentic behavior. You can able to control, and the control pane will give you all the guardrails, the policy which you can apply to the agents. You can have a monitoring and the observability around it and then apply the kill switch when it is going beyond. I think that learning, you can infuse it and scaling further to this.
John Furrier
>> I mean, basically risk management's built in.
Nikhil Gauba
>> Yeah.
Nikhil Gauba
>> Exactly.
Nikhil Gauba
>> I think adding to that, is it creating a value? Is it creating an ROI? So agents are failing because there is no measured KPIs on that, and then has it the right foundation? Those two are more important for me.
John Furrier
>> I wrote a post last night that said that Boomi's entering in the rise of the control plane in Chicago, little bit of teaser for the keynote. Read his hand blind a little bit there, but I interviewed him earlier. He didn't reveal anything. Steve didn't tell me anything, but I put it together. But I did make a comment. I want to get your reactions. I think you hit a great point there. Most organizations think chatbots, they came in from the chatbot side, and yeah, they're putting out some wrapper stuff and you get a prompt, you get in some sort of action or outcome experience. But the people that are designing their businesses, redesigning their businesses are looking at things much differently. Architecture, governance, and ROI, that's the new execution playbook. What's your thoughts on that?
Nikhil Gauba
>> Yeah, and the cost of operations because tokens are being involved, so there has to be a cost management part to it. So you're right, I mean, these are the basic things. Whenever we think of an agent in a business process, it has to be there.
John Furrier
>> All right, let's talk cost. FinOps is just now getting standardized in the cloud world, but we're beyond the cloud.
Nikhil Gauba
>> Exactly.
John Furrier
>> Remember, cloud had cost optimization, the whole FinOps category emerged from cloud. Now you got AI, it makes cloud look like it's settled. I mean, AI is probably a 10X more complex cost equation. Thoughts on that where you see the, I don't know what to call it. It should be called token economics.
Nikhil Gauba
>> AIOps.
John Furrier
>> FinOps for AI, what would that look like?
Nikhil Gauba
>> Yeah, I mean-
John Furrier
>> How should people frame at least their mindset?
Nikhil Gauba
>> Token management and cost management as a capability has to be at enterprise level, and then how do we monetize the agents based on that cost has to be in the enterprise. It is not just cost management, but even the monetization play for the business line. I mean, it is very foundational. Many enterprise are not looking at it now, but like the cloud, when they see the cost exponentially increasing, that has to be an important part.
John Furrier
>> It's funny you bring that up. I was talking to a chief human resource officer. As you said, there's already internal politics around who has more agents in their organization. And then there's also an effort to evaluate the performance of those agents, and then so there's a reorg mindset going on. That's a technical policy. How do you guys view that? Because I think you're right. We want to have agents under organizations, and hey, I got a ton of agents, some are doing work, idle, are not performing. Potentially, okay, you could have more agents in your department. Joking aside, but that's happening.
Nikhil Gauba
>> Exactly.
Nikhil Gauba
>> I think the way you can able to control it is making sure the agents are very contextual so that the number of tokens you're limiting. And that's why I think the prompt engineering is coming also as a handy part so that you can able to train your agent how those prompt will be structured. That's how you're controlling. I think FinOps, one of the key point is early detection of those wrong prompts, and then not hitting all the way to the token or the LLM.
Nikhil Gauba
>> Right. I mean, even the frontier models, which are large language models, they are expensive. We are also recommending small language models which are most cost-effective and enterprises can use in more use cases, which they don't need to use LLM.
John Furrier
>> And it saves on compute and GPU costs.
Nikhil Gauba
>> Exactly. Companies are moving towards using SLMs and managing costs for the tokens and how they do the prompt engineering to Yogesh point is there is a lot of enablement to be done and we can save a lot of token costs by how you tell the functionality should be done by an agent.
John Furrier
>> And you guys are basically talking about why I brought up architecture. We're in an architecture strategy game, not IT. I mean, ultimately, because it's a systems, it's not... It's a system effect.
Nikhil Gauba
>> Yeah. I mean, if you look at the agentic architecture, there are seven or eight layers. It's not just you have built an agent. There are so many layers which goes beyond that so all of those layers have to be set properly with the right guardrails. It includes how do you write prompts? What are the prompt injection in that? How do you control the data which is coming out of your assets which are connected by Boomi? So it's not just agents and doing something, but it has to be right across all the layers.
John Furrier
>> And you guys-
John Furrier
>> Also like to add what Steve mentioned in his keynote, the connect. If your connect is mature, then you can have the right context. Really, the agents can hit the right context to those APIs.
Nikhil Gauba
>> Activating your data.
John Furrier
>> Activating your data.
John Furrier
>> I mean, I do like the concept of connect everything, get the contextual data and understand and route accordingly. I love the new acquisition and then the control point. I mean, that's the three Cs right there.>> Yep.
Nikhil Gauba
>> Yep.
John Furrier
>> All right guys, final question, I guess I'll maybe do a round table on this one. Eddy, you're going to weigh in from the Boomi side, but on the Cognizant side, people want to know what this event means because it's an inflection point for Boomi because they got a great track record and integration. I mean, IPaaS segment people that know that, know that's a lot of work that's gone in and they certainly have the bar high on doing that. But now it's shifting to real time, everything's being integrated in real time. When agents are being redeployed or policy is being set, you got to have, they got to bring that expertise to the table. We saw a peek of the future on the keynote. What's the new Boomi look like? If someone asks you, what is the new Boomi opportunity? Knowing what you guys know, because it's a new Boomi. It's a bigger market TAM, bigger use cases, more horizontal use cases and all that integration skillset comes to the table. What's your thoughts on this?
John Furrier
>> I see today the launch was Boomi Connect. I'm yet to see that, but I think already the enterprises where we were helping earlier when we were dealing with them, it used to take weeks and months to really develop those, but then Boomi Enterprise already gives a lot, many connectors and a lot of Boomi AI components are making already the ease to make those things within a week or within a day.
John Furrier
>> So faster integrations-
John Furrier
>> And now within minutes, faster integration.
Nikhil Gauba
>> I think more than that, it's the personas and stakeholders we are targeting. It'll be totally different now. We'll be talking to CFOs, business stakeholders, maybe sales and marketing. We are influencing and creating agents for business outcomes, not just IT outcomes.
John Furrier
>> It's a business conversation you say.
Nikhil Gauba
>> It's a business conversation.
John Furrier
>> Boomi up the stack.
Nikhil Gauba
>> It's a larger game, more money for us. And go Boomi and Cognizant.
Nikhil Gauba
>> Go Boomi.
Eddy Montalvo
>> In a good way, right? I think ultimately, I think it's such an exciting time for Boomi because we're putting ourselves at the center of the transformation efforts that we can do for our clients. So when we talk about activating the data, right? It puts you at the forefront of every conversation with every business stakeholder and say, "How do you really want to go about this transformation?" And luckily for Boomi, its toolbox has put us in the front and center of that problem statement and how we really go solve that opportunity.
John Furrier
>> Yeah. I mean, my big takeaway is governance is not just compliance anymore, it's operations, execution, risk. So system of execution will be the agents.
Nikhil Gauba
>> And that's very unique positioning of Boomi. They have taken the AI governance side of the house, which is the key for making your agentic journey very successful.
John Furrier
>> Show me people that like to work on governance. Hey, who likes governance?
Eddy Montalvo
>> We talked about it a few times. Going to do it in a federated model, because we know they're going to use a lot of different tools, right?
Nikhil Gauba
>> And building trust.>> Building trust, yeah.
John Furrier
>> It's like developers, they don't want to do that, they don't want to debug anything, but they have to. But compliance, if done right, makes everything fly.
Nikhil Gauba
>> Right.>> Yep.
John Furrier
>> Guys, thanks for coming on in theCUBE. Really appreciate it. Looking forward to getting more insights down the road. Appreciate your time.
Eddy Montalvo
>> Yeah, likewise, thank you.
Nikhil Gauba
>> Thank you so much. Thank you.
John Furrier
>> I'm John Furrier here at theCUBE. Again, governance, these new system architecture, architecture, execution, return on investment. That's the new criteria, not just prompt to get a little action and some outcome. The results are going to come down to revenue and ultimately enterprise wide scale. Boomi's going after it here in Chicago. Thanks for watching.>> Thanks, bye-bye.
John Furrier
