Exploring AI Agent Workflows with Jesse Shiah of AgilePoint
Jesse Shiah, co-founder and Chief Executive Officer of AgilePoint, brings two decades of industry innovation to the AI Agent Builder Summit. With a background in mission-critical systems, Shiah shares their vision for AgilePoint’s role in enabling resilient and adaptable enterprise architectures. This video, hosted by a principal analyst for AI at SiliconANGLE Media and theCUBE Research, explores orchestrating multi-agent workflows that optimize human and digital collaboration.
In this session, Shiah discusses AgilePoint’s approach to building agentic systems that seamlessly adapt to organizational changes by leveraging AI technologies. Key topics include the company’s history of creating resilient systems, the evolution from rules-based workflows to dynamic agentic processes, and the importance of holistic abstraction to harness the potential of AI. The dialogue offers valuable insights on managing AI agents and orchestrating workflows.
Key takeaways include the significance of integrating diverse AI agents for enhanced business processes and the benefits of AgilePoint’s AI control tower. Shiah emphasizes how this framework allows for real-time adaptation and democratization of AI without coding adjustments, thereby minimizing risks and fostering trust. The discussion highlights AgilePoint’s strategic positioning in the AI marketplace and its ability to integrate third-party agents seamlessly, based on the analysis provided.
Forgot Password
Almost there!
We just sent you a verification email. Please verify your account to gain access to
AI Agent Builder Summit, Encore Presentation. If you don’t think you received an email check your
spam folder.
Sign in to AI Agent Builder Summit, Encore Presentation.
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 this link to automatically sign into the site.
Register For AI Agent Builder Summit, Encore Presentation
Please fill out the information below. You will recieve 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 Agent Builder Summit, Encore Presentation.
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
AI Agent Builder Summit, Encore Presentation. If you don’t think you received an email check your
spam folder.
Sign in to AI Agent Builder Summit, Encore Presentation.
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 this link to automatically sign into the site.
Sign in to gain access to AI Agent Builder Summit, Encore Presentation
Please sign in with LinkedIn to continue to AI Agent Builder Summit, Encore Presentation. Signing in with LinkedIn ensures a professional environment.
Are you sure you want to remove access rights for this user?
Details
Manage Access
email address
Community Invitation
Jesse Shiah, AgilePoint
Exploring AI Agent Workflows with Jesse Shiah of AgilePoint
Jesse Shiah, co-founder and Chief Executive Officer of AgilePoint, brings two decades of industry innovation to the AI Agent Builder Summit. With a background in mission-critical systems, Shiah shares their vision for AgilePoint’s role in enabling resilient and adaptable enterprise architectures. This video, hosted by a principal analyst for AI at SiliconANGLE Media and theCUBE Research, explores orchestrating multi-agent workflows that optimize human and digital collaboration.
In this session, Shiah discusses AgilePoint’s approach to building agentic systems that seamlessly adapt to organizational changes by leveraging AI technologies. Key topics include the company’s history of creating resilient systems, the evolution from rules-based workflows to dynamic agentic processes, and the importance of holistic abstraction to harness the potential of AI. The dialogue offers valuable insights on managing AI agents and orchestrating workflows.
Key takeaways include the significance of integrating diverse AI agents for enhanced business processes and the benefits of AgilePoint’s AI control tower. Shiah emphasizes how this framework allows for real-time adaptation and democratization of AI without coding adjustments, thereby minimizing risks and fostering trust. The discussion highlights AgilePoint’s strategic positioning in the AI marketplace and its ability to integrate third-party agents seamlessly, based on the analysis provided.
Jesse Shiah, co-founder and chief executive officer of AgilePoint Inc., joins theCUBE’s Scott Hebner at the AI Agent Builder Summit to discuss how agentic systems are transforming enterprise architecture. With a background in mission-critical systems, Shiah offers insight into building resilient frameworks that support dynamic human-digital collaboration.
The conversation explores AgilePoint’s AI-enabled orchestration platform and its evolution from rules-based workflows to adaptable agentic models. Shiah details the company’s AI control tower, which e...Read more
exploreKeep Exploring
What are the possibilities and capabilities of agentic workflows and how are they changing the way organizations operate?add
What is the concept of closed-loop agentic systems and how does it differ from traditional AI agents in the marketplace?add
What are some challenges that executives have mentioned regarding AI adoption and development costs, and how can these challenges be addressed through abstraction layers such as the AI control tower?add
What are some advantages of democratizing agentic AI for business people?add
What is the importance of governance in the implementation of technology experiments in real business settings?add
What capabilities does AgilePoint showcase in enabling cross-platform governance for AI agents in customer issue resolution workflows?add
>> Hello, welcome back to the AI Agent Builder Summit here on theCUBE. I am Scott Hebner, your host and the principal analyst for AI SiliconANGLE Media and theCUBE Research. Thank you so much for tuning into this session, one of a dozen at the Summit. Our objective is to find high value signals within the noise of the marketplace and to amplify the insights and best practices of industry innovators gain through real world experience in building AI agents and agentic systems. In today's session, we're going to elevate above the AI agent level to explore how to wire together, orchestrate and govern multi-agent, multi-vendor agentic workflows, which autonomously optimize based on real time feedback and collaboration among human workers and agent coworkers. Really cool stuff. While all monitoring and enforcing security and policy safeguards along the way. We are moving beyond the era of rules-based static workflows into a new world of agentic workflows, which understands an organization's evolving needs and the changing conditions inherent in every business and dynamically adapting as needed. And even better, you can hire agents, workers, if you will, from anywhere you want. Yep. Employ best of breed digital workers without constraints. It's all powerful stuff and this is becoming more and more of a reality every day that goes by as you will soon see today. I'm thrilled to be joined by Jesse Shiah, an inspiring leader who obsessively has been driving insights from his customers to help them be successful. He understands their aspirations, their needs, and their practicalities. Jesse is the co-founder and CEO of AgilePoint, which started over 20 years ago with a simple vision, build resilient and future-proof enterprises. So welcome, Jesse. I can't tell you how excited I'm to have you here and to see you in person.
Jesse Shiah
>> Absolutely, Scott. Thank you so much. So great to be here.
Scott Hebner
>> Yeah, we had a really good discussion on the pre-summit session, which I encourage everyone to go out there and make sure they see. I think that was a really good overview of the marketplace and where this is all heading and what you guys are trying to do differently than others. But why don't we start again and just kind of summarize the vision that you have at AgilePoint and what you've been trying to accomplish over the last many, many, many years.
Jesse Shiah
>> Absolutely. I'll give a short version today. So we are talking about agentic AI. As a matter of fact, our journey of this actually started 20 years ago, but part of that was actually influenced by my research at school was about how real-time systems means that the systems of running the mission-critical environment needs to respond to unpredictable events and about how can the system adapt itself to respond to that. And it is very similar to what we talk about a agentic today, I mean, but in a way, in order to make sure that even you make the change to the system, you still have the trust, you still have the confidence. So how can you do that without involving changing code? That including generating new codes or change the existing code. So that was actually architectural that we were thinking about even back in 1990, in the middle of 1990s. And then we saw the dot-com crash in 2000. We learned that dot-com is all about supply chain integration and you want that agility resiliency of the supply chain, end-to-end supply chain connectivities. And that's why I saw that come crash and I realized the market was not able to deliver that aspiration. And that's really what kind of dawned on us to say, "Wow." So we create the architecture can actually work in the business domain as well. And if one day AI is here, we are going to need this architecture. So we started out 20 years ago building this.
Scott Hebner
>> So it's been 20 years in the works. AgilePoint today, I think your big focus is these closed loop agentic systems, which we're going to get into here in a little bit more detail. And I think along the way of this long journey you've been on and now with the advent of AI and then now agentic AI, you are incredibly well positioned. I think you have a lot of really core differentiators, at least how I see the marketplace. But we'll get into a little bit more detail. Just tell me about how you see this all coming together, sort of the vision into the future, if you will. And to do that, let just bring up a chart that we have here.
Jesse Shiah
>> Absolutely,
Scott Hebner
>> Go ahead. Why don't you kind of describe what you have in mind here as we go forward?
Jesse Shiah
>> Absolutely. I mean, we talk about AI and then we talk about AI agents. And I think our focus from day one when we looked at that, as I mentioned that for example, even the supply chain is end-to-end. So it's not specific to one applications or one functional areas because as a matter of fact, in the organizations, if you take the 80/20 rule loosely, about 20% of the applications are actually that cross-functional end to end and the rest of 80% maybe a lot of application for functional specific and tied to different platforms. But the 20% of the end-to-end actually drives probably 80% or more of the business outcomes. So when we look at the agenda today, you see the three levels in this chart here and there's AI agent, and I think that's where the most of the market today are doing right now. They consider this agent can help me make this task or decision better. And you see a lot of automation , because the architecture they built was not based on the architecture that we actually set out to build. It's completely abstracted, composable. So the architecture is actually still turn automation intent into executions. So because of that, it's very easy to just boil down the AI agent to a specific task. And while it does provide incremental productivity gains, however they impact the AI agent now being limited to each silo's tasks. So I think that's where we, from the chart, you can see you can go from AI agent to agentic, agentic is where you have multiple agent now. But what we're seeing a lot more is in a specific platform if a multiple agent can work together. But we think when agentic will become the next big thing for AI is really when you can get to that real-time agentic systems being a system that applies that end-to-end. Okay. And not just end-to-end actually multiple of them can actually coordinate among each other in real time.
Scott Hebner
>> Yeah, that's the closed-loop thought, right?
Jesse Shiah
>> Exactly.
Scott Hebner
>> Yeah. And it's cross-business and all that.
Jesse Shiah
>> Exactly.
Scott Hebner
>> And you're right. By the way, we plan to have multiple, a series of these summits on agentic AI. You notice the name of this summit is the AI Agent Builder because that's where it all starts. You need to have a marketplace of agents to do tasks as you said, or to become to mirror someone at a co-worker if you will. It's another whole thing to be wiring these together into workflows.
Jesse Shiah
>> Absolutely.
Scott Hebner
>> And that's where I think you're coming at with what you've learned over the last 20 plus years. And you've come up with this notion of a holistic abstraction. So can you just explain what that means and maybe how it compares to other agentic platforms, ServiceNow or Salesforce for example?
Jesse Shiah
>> Yeah, I don't want to compare to other platform, but it's better that to share the perspective. And as a matter of fact, I think there's, take one example. I think maybe over two months ago or two and a half months ago, Microsoft CEO Nadella had a podcast and he mentioned about eventually all the business logic will go to AI agents. But you think about it, all the business logic, it doesn't mean that it doesn't say that the logic of the HR system, the logic of your CRM, the logic of your ERP is all business logic. So if you think about that, that means that you want agent to be able to detect, be able to analyze, detect, predict, and then we will take actions autonomously. But the most important thing you'll be driving, you'll be driving the business goals continuously and continuously optimize that. And to make that possible means that from LLM being a lot of content, a lot of words, you want to turn into actions. But now it's how can you turn that into actions? And then also carry out the action with trust. That's really key. That's really key. And that's why we talk about in architecture, that's not generating code, that's not modifying code from day one. And in this case here, if we want to enable agent can actually adapt your business logic of your applications, especially we say the most valuable use cases, that end-to-end orchestration. That means that the application itself, your automation, your orchestration has to build in a way to enable agent to change them, change the logic, but without involving code changes. That's what we mean by holistic abstraction. So today in the AI agent space, now, I think people are realizing we need more abstraction. So the MCP, I'm sure you know about MCP, but that's not enough. Basically we need to do abstraction on both ends. You need more abstraction on the AI side, on the agent side, the abstraction also needed on the applications, automations that it will be governed, will be adopted by AI agents. So that's what we mean that it has to be, the whole spectrum has to be abstracted.
Scott Hebner
>> So with our friend George Gilbert, who I know you know.
Jesse Shiah
>> Yes.
Scott Hebner
>> His notion of the harmonization layer-
Jesse Shiah
>> That's exactly. That's exactly.
Scott Hebner
>> Is very related to what you's talking about with holistic abstraction. So that makes a ton of sense. And let's take a look at a video that I know you guys have that we're going to kind show.
Jesse Shiah
>> Absolutely. So let's take a show video just kind of just shows how the platform, how do you do that no-code composition be able, very easy to create an end-to-end orchestrations and automation.
Scott Hebner
>> Okay, let's do it.
Jesse Shiah
>> Let's start off with understanding the core AgilePoint functionality where 15 process patterns and more than 1200 building blocks combined with AgilePoint's codeless architecture will not only simplify the creation and maintenance of process automations of any complexity, but will also set the foundation for the use of complex and potentially dynamic agentic processes on top of the IT investments that you've already made. 120 systems have been abstracted into reusable and configurable connectors that won't require expert skill sets in order to use so that I can harmonize data across systems wherever it resides. And to give you a brief sense of how this would work, let's drag and drop an activity from my CRM system where you'll find Wizard based configuration to guide me towards my intended outcome. I'm working with a customized object, but AgilePoint provides visual mapping to pass data between systems just like this. You're seeing a responsive HTML five interface builder where I build and consume on any device or orientation, even if offline, to build the interface, there are more than 75 widgets each with configurable validation and a visual multi-level rule builder with full support for event-driven rules. And lastly, an optional and business data store so that as your applications run, data is collected and ready to train AI.
Scott Hebner
>> I mean, it brings it to life, you can adapt very quickly. You don't have to send it off to be recoded and you don't have to-
Jesse Shiah
>> And eventually, you have all this agentic application and they will be able to talk to each other. And so eventually you get to a no-man business.
Scott Hebner
>> So your agents are going to help people accomplish tasks, their co-workers, and then you put them into agentic systems, which then you get multiple agents, co-workers working together.
Jesse Shiah
>> They can work to... Exactly.
Scott Hebner
>> And then where you're taking this is then you have a more closed-loop agentic process-
Jesse Shiah
>> Each one of them. Exactly, exactly.
Scott Hebner
>> It's adaptable, it harmonizes all your applications to your data and allows you, because of that, to integrate multiple processes, link processes together.
Jesse Shiah
>> Exactly.
Scott Hebner
>> And you start... yeah.
Jesse Shiah
>> Can bring in technology platform in and out as .
Scott Hebner
>> So if you're integrating multiple businesses or processes, let's call it that, tell me a little bit about the orchestration, because I think what you're saying is you can do it end-end across agentic workflows, not just one. Did I get that right?
Jesse Shiah
>> Exactly right. I mean, I think that's what we talked about, the abstraction, the harmonization across the technology stacks. So today we have abstracted more than 120, that's most popular systems already. And the framework enables additional platform to be, you can bring in new technology in, apply the same abstraction. And this is very different from how you go to most of the IT environment today, it's very platform-centric. You go to a lot of companies today, they have a CIM platform, for example, Salesforce. They have a team dedicated for Salesforce development. You go to maybe ERP, SAP, they have another team dedicated for that development. So the way our operation today is, it's just a lot of this platform silos, but then the end-to-end is being created by exchanging data among them. So we are talking about in this case, business usually cannot touch that. It's not abstracted enough for business to do that. So we take a very different view from day one. We say, okay, you as an organization, what system do you have? What asset do you have? Through this end-to-end abstraction and harmonization, we elevate them. All the different platforms and system now speak the same language. Same language we call metadata, it's metadata.
Scott Hebner
>> Harmonization.
Jesse Shiah
>> Exactly, that's where you got the harmonization. But when you get to that level now, so the business now can actually describe that end-to-end, the model. It's a model, it's a model-driven, model-based involvement. I mean, business are being so used to describing the business requirement in different diagrams. I just said Microsoft Visio, if you remember. And so in this case they can just describe the model, but what they are actually describing how the different systems should work together in the same language.
Scott Hebner
>> So let me try to put the pieces together so far. You can employ agents, individual agents that can help people be more productive, task oriented, and all that. You can wire them into a workflow-
Jesse Shiah
>> Yes.
Scott Hebner
>> To do more complex things, more sophisticated things that take more than one agent. Got that. Then to really create though a process, most processes, particularly if you're going to integrate multiple processes that are across businesses, that is largely implemented by applications. Those applications are probably like Salesforce are going to start to surface their agents to deal with that. What you can then do is you're harmonizing all that. You can take all the agents from those applications and you mentioned all these, the people in IT have all these different silos. They're going to be surfacing agents, but your platform's going to allow them to wire it all together. That's the abstraction layer. The orchestration is how they operate-
Jesse Shiah
>> Correct.
Scott Hebner
>> In a real business process.
Jesse Shiah
>> Exactly.
Scott Hebner
>> Okay. And then the next thing I believe you do is allow business users or certainly not the heavy duty developers in IT to actually make changes with this notion of a model-driven updates, right?
Jesse Shiah
>> Correct.
Scott Hebner
>> So can you explain that, the run times?
Jesse Shiah
>> Yeah, definitely. So from agenda perspective there also, we heard a lot of executives tell us that AI is so expensive. AI development is so expensive. And the other thing is, and they found out is the adoption of AI have to be slowed down, because they don't have budget to train their business users. But if you think about it, if you boil down the agent right into your business process for example, then you put it inside right in front of face of the business users and they must understand what that is so that they will not make mistake and incur risks. So we actually through the abstraction, so not only on the, you have the end-to-end harmonize, you can build that end-to-end orchestration, the abstraction. Even for when you want to operationalize AI, we have another abstraction layer called AI control tower layer. We developed that back in 2017. So that's where the old agent can go to the AI control tower layer. So what that means is you can operationalize AI agent without physically change the business process. So that provides a huge benefit. Now, business don't have to understand AI for you to operationalize or adopt AI. And the AI developers now, they do not have to understand operations. The AI control tower brings them together. That's where you operationalize, you govern, but you also democratize AI.
Scott Hebner
>> So is it sort of conceptually, metaphorically I should say, the concept of your HR department where all the agents are up there, they all have the ability to do work, and they're available for business users to then take and make modifications to or implement the processes and they can do it all through your platform without being as mere mortals, not being heavy duty technologies writing code.
Jesse Shiah
>> Exactly. So that layer, if you think about it, that layer now, because it's not hardwired or associated with the process, so when need to make change now, when you change the layer, all the process that are related, they all received the new change. It's not about technically can you do something, it's all about manageability and also and that directly relate to just how flexible. We talk about the real-time agentic systems, without this architecture, it's not possible.
Scott Hebner
>> Yeah, and I can see what you're getting at because there's a couple attributes that are, like I mentioned before, you spend so much time with customers and understand the practical nature of what they're trying to do is their IT departments have a lot of technical debt. There's not a lot of free time to be off building things quickly. With this, the business users are in more control, but within an architected framework on how you wire these things together. The agents provide the productivity and help with decision intelligence and explainability to help these workers be more productive and to work with each other. And you just simply put is, you're able to do more than you would otherwise be able to do, one. Two is when things change, when needs change, you can implement those changes quicker, because you don't have to go to IT.
Jesse Shiah
>> The business can do that. So basically the system enable IT to set up to extend, but however now you enable business now to be able to actually innovate and respond to business requirement changes without having to go back to IT for every single change. I always say that if we have to go back to IT, we put IT in a position that is a slippery slope, because it never ends. So it's more like you set up an environment, but this environment provides the governance, the securities. And so the business can actually innovate without incurred risks.
Scott Hebner
>> It's like business conduct guidelines.
Jesse Shiah
>> Exactly.
Scott Hebner
>> The policy and all that. So you're democratizing agentic AI for business people.
Jesse Shiah
>> And so you can adopt faster and not only you can adopt faster, you are able to scale faster. And the other thing is you can also adapt to change at scale
Scott Hebner
>> And you're able to integrate across processes and workflows more seamlessly. And these agents are just abstracting how you can do all that and how you wire them together. One of the things I find really interesting about what you're doing is the notion of any agent. To me it's just like, "Hey, you can go employ anyone you want. It doesn't matter what their background is, where they came from. You're able to go out and get best of breed agents and employ them within your business and your workflows and all that." I find that really interesting and I have read many, many studies. McKinsey for example, had one not too long ago that pointed out a majority of businesses are going to acquire agents and customize them and then integrate them in. They're not going to try to build all their own agents, they'll build their proprietary ones, but for more common things that can be customized, they're going to buy. So it's going to be a mix.
Jesse Shiah
>> Of course.
Scott Hebner
>> And with your platform at AgilePoint, it really is agnostic to the source of the agent. You can take agent from pretty much anywhere and do everything you've talked about. Is that-
Jesse Shiah
>> Absolutely. Absolutely. I mean, like I said, we just carry the same philosophy. Before agent got here, we say, "Okay, how do you build the end-to-end automation orchestration in the way it's highly adaptable?" Before agent got here, we enabled business to do that. The business user are able to respond in real time to that. And now we can say that, so your platform, your ideas have become more pluggable. But with the AI agents, it's the same philosophy. Yes, used to be I think ERP vendors kind of say ERP is the only system you need, but in the lab it proved that that's not enough. With the agent, it's just going to be even more versatile. So organization must be able to support that. Not only you create your own agents, but also be able to bring in third party agents, off the shelf agents. But it's even more important just the speed of the technology change in the AI space. We must make those completely pluggable. And so the whole architecture AI tower layer, we talk about the framework was designed to do exactly that.
Scott Hebner
>> And by the way, one thing I have been following is the emergence of all these AI agent marketplaces as being different from the gen AI plugins. Real marketplaces where there's business agents that you can go out and buy and download and customize and all that kind of stuff. And I think as that takes off, it's going to drive this notion that you have to be able to wire them together and not be tied to a particular development methodology or set of codes. So I think that's really important.
Jesse Shiah
>> Exactly. And I think, I mean, not only you can bring them in, it's really about being able to govern them. So there are a lot of agent-
Scott Hebner
>> As a whole.
Jesse Shiah
>> Yeah, as an agent, build a framework out there, they can build agents. We're just finding in a space right now, I think there's too much technology experiments. Technology can do it, but can you put it in a real environment, in the real business settings and you have the confidence? So the governance is truly important. Because we already have supporting global enterprise as early as since 2006, the user to do this mission-critical end-to-end orchestration that require a lot of governance. So we have that layer governance framework and that's also available now when you go to the AI control tower framework. And that's why if when you're plugging a third party agent and they don't have a strong governance, you can actually leverage AgilePoint's governance for it. So as a matter of fact, why we have a short video, we can take a look, seeing that it looked very simple. Let's take one third party agent like LangChain, and then there's another let say AgilePoint process. And this is really interesting. You see AgilePoint process model, you only see one activities. But I think the use case is like customer feedback. Customer may ask any kind of questions, but how can you dynamically produce the content for the customers? So LangChain will be receiving that, analyze that, do the planning, say this is what we should do. And now you can see how AgilePoint agent now will actually leverage the dynamic composition of the process. And that process where actually tied to using your actual systems, your actual data, provide us a tool to execute that. So because of that, now your agent are being trained not just by some public data, it's actually your systems data coming from your real systems, running your real business operation. That's how you'd be able to really contain the hallucinations, but also enhances trust. We can take a look at the show video, I think.
Scott Hebner
>> Let's do that.
Jesse Shiah
>> Yes.
Scott Hebner
>> Let's take a look at the video.
Jesse Shiah
>> In this demo we'll showcase how AgilePoint enables cross-platform governance for AI agents using a customer issue resolution use case. You'll see how it creates complex multi-vendor agentic workflows all without hallucination. So a customer will enter their feedback into a simple application, which is going to be handed over to LangChain, our planning agent and dynamically collaborate with AgilePoint agents to execute that plan. Moving into the application, the customer is having technical issues with the products that they've purchased, all context that AgilePoint is aware of. Next, let's illustrate how AgilePoint brings observability and explainability of agent decisions and actions. In green, you're seeing the same simple three-step process from our designer view and in yellow, the results of our planning agent. Three dynamic process compositions invoked by LangChain, all of which have come from a vetted and approved catalog of processes. Now this customer faced a technical issue, so our first AgilePoint agent has created a support ticket. Our second agent, having the context of customer purchases has assembled a personalized troubleshooting guide and finally a personalized article, all of which have been delivered to the customer via email. In closing, let's kick off one last instance to see some variation in planning and execution. Rather than a technical issue, this customer will face a billing concern. So as you're seeing when the issue changes, so does our planning and resolution. So not only is AgilePoint bringing the tool sets for our planning agent, we're providing the observability and explainability and all of this is captured in AgilePoint's audit trail for ongoing model retraining.
Scott Hebner
>> Again, seeing is believing. It makes it really easy. So you can take crew AI agent, LangChain as you showed, you can wire them together.
Jesse Shiah
>> Yeah. If you notice that LangChain was actually one agent, there's only one AgilePoint agent, but did you see that extended, all the yellow? There are three branches. So LangChain suggests there are three different options that could actually be, say a supply chain, self-healing supply chain scenarios. Now you suggest that, okay, now we're going to have the three new routing options. And all those were dynamically generated by AgilePoint, but there's no code generated, there's no code being changed. And that's a power.
Scott Hebner
>> That's why you can plug it in.
Jesse Shiah
>> Exactly. That's a power. But it actually drives the business, it drives your business operation now and without introducing some unpredictable risks.
Scott Hebner
>> And that's what the control tower is all about.
Jesse Shiah
>> Yes.
Scott Hebner
>> As you know, the theme here is trust. Trust is the currency of innovation. No trust, no ROI. That's what the control tower is all about when you're wiring together all these multi-vendor agents. Before we run out of time, how about a real-world example?
Jesse Shiah
>> Sure. So we actually just were discussing with one of the global SI and for one of their very large customer in the life science pharmaceutical industry for example, they talk about, I just mentioned about the example of a self-healing supply chain. So in that case, so you have a supply chain, the end-to-end supply chain. And kind of interesting that what I mentioned that when we got started, when we saw the dot-com crash, it was really company say they could not realize the aspiration of that resilient supply chain integration. So in this case here, the same thing. So if you think about supply chain system that may involving maybe advanced planning systems, you may have the ERP, your CRM, maybe your customer support systems. And the thing is, like I said, when you detect the demand pattern change or maybe the weather or maybe the shipment delay that subject to compliance and SLAs, but how can you dynamically reroute? So we are talking about that's making one the next best action. But the example we just show you, the LangChain one you just saw, for example, that could be not just the next best actions, the next best execution, but that execution could be actual sequence. And that is actually a really good example of talking about, I mean, just kind of imagine what this could mean now for business.
Scott Hebner
>> And each part of it understands the other part, one part changes. And again, you're able to wire together best of breed in your application set, harmonize it. It's really great stuff.
Jesse Shiah
>> I mean, we actually did our webinar in January last year. So we called, this is a agentic enterprise automation fabric. It's matrix, it's a matrix. You have automation for functional, you have your end-to-end. Each of them individually, agentic driven and doing optimization. But however, among each of them they'll with each other as well.
Scott Hebner
>> Really great stuff. And we're definitely going to stay in touch here. I know we're going to talk more and hear more from you and maybe we even go to the next level of detail.
Jesse Shiah
>> Absolutely.
Scott Hebner
>> I really appreciate you being here. This has been a fascinating conversation and congratulations on the progress you're making.
Jesse Shiah
>> Thank you so much.
Scott Hebner
>> You bet. And for you all, please make sure you visit the dedicated portal out on the AI Agent Summit website for AgilePoint. When you go out there, you get this video, the pre-summit video, and get some access to some of some really great reading materials about Agile Point and their point of view about what's happening in the marketplace and what they're able to provide. And of course, visit agilepoint.com to dive even deeper into the topics we discussed today. Finally, you can access the entire AI Agent Builder set of videos out at thecube.net or on our YouTube channel. Please share those videos, the articles and the clips of all the sessions, and in particular this one with your colleagues and on social media. We'll see you real soon again. Thank you. We are the leader in enterprise tech news and analysis. Bye for now.
Scott Hebner
