Certinia at Dreamforce | Tyler Philpot, Thunder

Clips
More from Certinia at Dreamforce

Tyler Philpot

VP, Solution Architecture

Thunder

play_circle_outline Thunder as Salesforce Partner: Aligning Pre-Sales to Delivery for Certinia PSA, Sales Cloud, Slack, Veda, Agentforce/AIforce Integration

play_circle_outline Deterministic Knowledge Graphs Replacing Probabilistic LLMs to Capture Continuous Project Context from Presales to UAT

play_circle_outline Reliable Clean Context from Drive, PSA, Slack, Docs, and Meetings to Prevent Overfitting, Cut Costs, Improve Customer Experience

play_circle_outline Economic impact: potential shift to outcome-based pricing, balancing efficiency with margin protection

Info

Tyler Philpot, Thunder

Tyler Philpot

VP, Solution Architecture Thunder

search

Scott Hebner

>> Hello everyone, welcome back to theCUBE's continuing coverage of Dreamforce 2026. And I'm Scott Hebner, I'm the principal analyst for enterprise AI at theCUBE Research and SiliconANGLE. Thanks so much for tuning in. We continue our conversations here at Dreamforce, and I'm excited to introduce Tyler Philpot, who is the Vice President of Solution Architecture at Thunder. So appreciate you taking the time to be here.

Scott Hebner

>> Thanks for having me.

Scott Hebner

>> Yeah, so why don't we start for our audience, just explain a little bit of what Thunder does and what your specific remit is around solution architecture.

Scott Hebner

>> Thunder's a Salesforce partner. We specialize in implementations and managed services across a variety of different clouds. And my role specifically as VP of Solution Architecture is that I do a lot of the pre-sales, scoping, estimating. Really making sure that our customers' journey is taken into account during the sales cycle so that we can make sure that we're targeting the program and make sure that we're delivering the desired business results and ultimately setting up projects for not just success but continued success time and time again.

Scott Hebner

>> So what do you think of the show so far, in particular Marc Benioff's keynote?

Scott Hebner

>> So far so good. The keynote was great. My favorite part was the walking and talking and the banter between the speakers there.

Scott Hebner

>> And it wasn't lost on me that he was wearing a suit, which we had lost that in the IT industry for decades.

Scott Hebner

>> Oh yeah, he's bringing it back.

Scott Hebner

>> He's always out on the leading edge of doing things, right?

Scott Hebner

>> Right, right, for sure. He is a leader.

Scott Hebner

>> Yeah. So I also noticed that you guys are part of the advisory board for Salesforce as a partner. So that must mean a lot, right, in terms of your authority and how they view you?

Scott Hebner

>> Expertise with the platform, connections, access to a variety of different partner tools and resources. So it really helps us command the marketplace and make sure that we're doing right by our customers.

Scott Hebner

>> Right. And you were one of the early vendors involved with Agentforce,

Scott Hebner

>> right?That's right. Yep. We've got several forward-deployed engineers and that's a product that keeps on running hot. So we're happy to keep going.

Scott Hebner

>> Yeah. Now they're adding AIforce. What's your thought on that? That next layer on the

Scott Hebner

>> architecture?It's the natural next step. Everything's moving in that direction. So we're doing everything we need to do to make sure we're on the cutting edge.

Scott Hebner

>> So a lot of what you do is help connect the pre-sales information to the actual delivery.

Scott Hebner

>> That's right.

Scott Hebner

>> And help keep things in context, right? And that breaking point, what has traditionally happened that's been a problem?

Scott Hebner

>> I'd like to say that's a process that's flawless and there are no gaps in it at all. But we all know that's not true. So even happy path when we have meticulous documentation and we have an accounting of everything a prospect has discussed during the sales cycle, even if we were to transition that seamlessly to the delivery team providing the project, there's always the chance that the client themselves, they have mismanaged expectations internally as well. So that's one thing that we've seen material benefit from working with Certinia and the Veda Accelerate product is not only does it keep us accountable to our own notes, but it's also a great way to make sure that we have strict accounting of who said what in which meetings and what's in scope, what's not in scope, what's been deferred, because the executive sponsor and individual stakeholders a lot of times have different perspectives, and having that complete knowledge graph that's really painting a rich tapestry of who's managing what expectations is really key for us.

Scott Hebner

>> Right, so it's traditionally been a hard problem to solve, because you're dealing with human beings.

Scott Hebner

>> Exactly.

Scott Hebner

>> Now you have AI.

Scott Hebner

>> The human element is always a wild card, so AI is keeping us accountable and keeping everybody on the same page. And that's, you

Scott Hebner

>> know—As I mentioned before we got into our conversation here, I've been a huge proponent of the knowledge graphs and context spaces and things like that. But before we jump into that, tell me a little bit about your architecture, your stack, if you will, that you've built.

Scott Hebner

>> Yeah, well, we are longtime customers of Certinia. So that is our PSA of choice. That's where we manage our projects start to finish. Obviously, Sales Cloud and everything in between there. Slack is a key element of creating that engagement layer where all this information is surfaced and actioned. And then Veda is the new toy to play with. And that's really adding that additional AI layer to make sure that our PSA and our project managers and engagement managers, everyone's working from the same dataset and everyone is looking at proactive signals rather than just reactive responses.

Scott Hebner

>> And that's across, as you said, from pre-sales, absolutely.

Scott Hebner

>> We're able to manage the health and signals from opportunities and the sales cycle, all the way down to, hey, we're coming to UAT next week, what aren't we thinking about that we should be?

Scott Hebner

>> So, AI historically has been very probabilistic in nature. I think Marc Benioff mentioned that they're driving towards more deterministic, but I think a step on that journey is the notion of knowledge graphs.

Scott Hebner

>> Yes.

Scott Hebner

>> And relationships between entities and things of that nature. And you talk a lot about continuous knowledge graphs. Can you just give our audience a little bit of a sort of an overview of what you mean by that and then what it sort of does and how you're doing it?

Scott Hebner

>> Yeah, when it comes to knowledge graph, it really, I've heard it said that, prompt engineering was the term of art, but now it's really context engineering. That's really what matters most. And I can't agree with that more because, when we're thinking about just how we balance continuity from a presale cycle into a project with varying perspectives and inputs and outputs and decisions being made along the way. That knowledge graph is continuously being changed. So what was scoped, in presales may have changed 4 times by the time you get to UAT. So you need to make sure you're continuously updating and refining what that knowledge graph looks like and really take into consideration everything, all the discussions and inputs that it's been able to analyze throughout the project.

Scott Hebner

>> Is the context part of what comes from that knowledge graph? Or is that an additional layer, like what they do with Moonnox?

Scott Hebner

>> Yeah, the context, it's coming from a variety of sources. We have our Google Drive, we have our PSA data, we have Slack channels. All of this is creating this rich tapestry of context. It really comes down to which context supersedes the other. How do we avoid overfitting context to make sure that the knowledge graph is complete, accurate, but also void of superfluous noise, for lack of a better term.

Scott Hebner

>> So when you're talking to a client or a potential prospect, do you get into discussing the knowledge graph and the power of it? Is it resonating?

Scott Hebner

>> It is a selling factor for sure. Just knowing that we're AI-enabled and we have these tools at our disposal to make sure that we're able to run their project. More successfully and ultimately translate into savings for them, I think that means a lot. And knowing that they're not going to have to repeat themselves also, because I think that's a common complaint across clients, that, hey, we're doing discovery, we've got, 3 weeks of in-depth conversations with all these individual stakeholders, with the architects and, the engagement managers running the project. But you know what, I told your pre-sales guy that, a month ago. Why do I have to repeat myself? So a lot of this is able to, or enables us to be able to build on that knowledge and add to it rather than just rehashing it across meetings.

Scott Hebner

>> So in many ways, it's building institutional knowledge about the project, about the client.

Scott Hebner

>> Absolutely.

Scott Hebner

>> Yeah. It takes the more general purpose normalized data that you have from all your engagements and starts to build a knowledge graph that applies that to that specific project. That's basically the value of it, where in the past it was more LLM-based, probabilistic. Yeah. You weren't building institutional knowledge.

Scott Hebner

>> Right, right.

Scott Hebner

>> And then increasingly having context around that.

Scott Hebner

>> Absolutely. Yeah, it's been huge. So that speeds the delivery cycle, lowers costs, and provides a better customer experience as well. And that's one of those intangibles that I think really goes a long way in making sure that the consulting relationship is as smooth and seamless as possible.

Scott Hebner

>> And one of the themes here from Certinia is Be Certain, right? And the system of action and you need to be increasingly certain about the actions you take and how they affect the outcomes. And I would imagine there's a direct correlation or tie to the knowledge graph and the context, right? Because you can't make a good judgment and Be Certain about it and feel good about it unless you understand the context.

Scott Hebner

>> Just like analytics, garbage in, garbage out. You got to make sure that the knowledge is clean and specific and tangible and helps inform those AI agents to be able to make those recommendations based on good information.

Scott Hebner

>> And one thing I learned about you guys is a term that I haven't really heard out there, but I think it's super intriguing, which is a context engineer. Why don't you just talk? I think I got the idea here, but for the, you know, I'm assuming it's someone that, like you mentioned prompt engineering. Yeah, is this a new discipline, evolved discipline, or?

Scott Hebner

>> I think it's quickly becoming, if it's not there already, the newest term of art that people are really focusing on. It's all about establishing your pipeline to make sure that by the time you get to your agent and you start prompting it to do something, you want to make sure that by that point it has the relevant context. And I think that's one thing that we're challenged with is how do we provide it with the documents that we got from our partners, with the, you know, interviews and the discovery sessions we had with our prospects and all these other sources of input. How do we curate it in such a way that by the time I get to that agent, it knows, hey, these are the documents that are relevant. Here's the context that I should be drawing from when responding to that prompt. Too broad context, you get more hallucinations and more difficult answers. But if you get very specific and very precise with your context, you get better answers.

Scott Hebner

>> So as you lead the solution architecture for Thunder, you have context engineering activities underway? Absolutely. And that's an interesting approach because I have not heard that term before, but I think it's important because it is different. You're right, it is different than prompt engineering. That's very different. And it kind of takes this notion that generative AI, the LLMs, are sort of like the browsers of the early— it's the gateway into the world of AI. But to really be successful, you got to build institutional knowledge around your particular domain business.

Scott Hebner

>> For sure. Yeah.

Scott Hebner

>> And that's the knowledge graph, and you have the context coming from it, and you need the discipline to be able to build and maintain all that. And I think that's what you're getting at.

Scott Hebner

>> Absolutely. Yeah.

Scott Hebner

>> Interesting. So where do you think this is all heading? Just in following professional services pretty much this full year, and Certinia in particular, there's been a lot of innovation. I just think back to March, and then they got into agentic knowledge. They were talking about knowledge graphs, and they had the Moonnox, and now the record system of— not record, but System of Action. What do you think in 3 months we're going to be talking about? Today is sort of like knowledge graphs and context, do you think it's moving quick enough that we have the next big horizon here for what you guys do?

Scott Hebner

>> Yeah, some of the big advances as of late, we have MCP servers becoming more ubiquitous and more autonomous agents working, and then we have, you know, headless architectures where we're no longer bound to these UIs. So it's only a matter of time, I think, until you have projects being run semi-autonomously through these headless solutions. It's going to be a wild ride, that's for sure. I'm certain of that.

Scott Hebner

>> Yeah, well, there's been a theme in some of the interviews I've been doing about people making comments that AI is actually generating more jobs and more work, which is a good thing because you're able to do your projects quicker, more efficiently, lower cost. Which then incents the client to do more. So it's not actually eating away people's jobs in professional services. Is that your observation too?

Scott Hebner

>> I haven't seen it yet. It's been able to take a lot of the administrivia, the low-level work, and allow that higher-level strategic thinking to take place. And we do want to make sure that at least at Thunder, we're not getting into that role of just ticket taker develop the solution. But let's make sure that, hey, if we can send off the digital labor to build the solution. Let's spend our extra time making sure that we're building the right solution for the right outcomes and really making sure that everything is buttoned up that way. And I haven't— yeah, I lost my train of thought there. Sorry.

Scott Hebner

>> Yeah, no, I think there's just a lot of— you look on the broadcast news and all the fear about AI taking everyone's jobs, and I just don't— I personally, I was a CMO for IBM's AI business for a while before I retired, then I've been an analyst ever since following this, and I just don't see the massive reduction of jobs. If anything, they're shifting, to your point.

Scott Hebner

>> There's definitely shifting. I think the roles will continue to be there, so we're still going to need architects, developers, analysts on projects, but the nature and how they work with their tools and with their digital labor counterparts, I think that'll be exciting to see. Faster throughput, maybe deployed to more projects. So there's a lot of things that can come out of it.

Scott Hebner

>> Yeah, I think people should be excited about it. And if you're in professional services, it's giving you superpowers to be more valuable and to do more things and have a better next step of career moves because you learn how to use AI effectively, you're going to be a lot more valuable than someone that doesn't.

Scott Hebner

>> It's definitely going to be an impactful technology to learn. It's either you adopt it and you become a multiplier, or if you lag behind, options would be a little more limited, I would imagine.

Scott Hebner

>> Yeah, I think the context engineering is a good example of that, sort of a new discipline, because you still need prompt engineers. Yeah, right. So it's not like, they get replaced. Now, some of those might be dual jobs, but again, it's creating more and more opportunities. one thing I believe in, the reason why I cover professional services, I think it is, if you want to know what's going to happen in enterprise AI, you look at professional services as an industry, because you guys by definition are ahead of the game, because just by the nature of what you do, you build solutions. Right, right. And how AI affects you and what you're able to do and the value you get out of it, it's ultimately going to happen in the other industries. Yep. So what happens in professional services is going to then happen. That's why I've been watching you guys so carefully. And you guys have made the move as an industry from a technology discussion to economics and outcomes, which I think is really healthy. Yeah.

Scott Hebner

>> Yep. That's the trend for sure.

Scott Hebner

>> A lot of the other industries I haven't seen that yet. Which brings me to the last question, just to get your two cents on, is this move potentially to fixed-price bids, outcome-based pricing versus historically obviously not that way. It was based on billable hours and materials.

Scott Hebner

>> Yeah.what do you see happening now?

Scott Hebner

>> Ultimately, we want to make sure that we're driving consistency and predictable outcomes with our customers, but also, financial side of things. If you get twice as efficient and you're working on a time and materials basis, it means you get half as much revenue. So that's the mathematics of the situation. So part of this is definitely customer base and making sure that we're driving predictable results. But also how do we protect our margins at the same time and make sure that digital labor isn't cutting into our revenue?

Scott Hebner

>> Right.

Scott Hebner

>> Yeah.No, I think it goes back to that notion that you do it quicker, more efficiently, at a lower cost, you delight a customer, then they're going to want to do more. And there seems to be an endless appetite for the other thing.

Scott Hebner

>> Exactly.And as AI keeps speeding along in terms of innovation, there's only going to be new things to do. So it seems like a very healthy dynamic that's occurring. Yeah. We'll come and take down into the other industries, just sort of leading the charge.

Scott Hebner

>> So success begets more success. So we're excited to use our AI and delight our customers and make sure that they're on a long journey with

Scott Hebner

>> us.You get to be a guinea pig.

Scott Hebner

>> Exactly.

Scott Hebner

>> Yeah.

Scott Hebner

>> Yeah.It's fun. No, no dull

Scott Hebner

>> moments.All right. Great conversation.

Scott Hebner

>> Yeah.Appreciate you being here. Appreciate

Scott Hebner

>> that.Thank you. All right. Well, thank you all for taking the time to be here. We will continue our coverage of Dreamforce 2026. Stay tuned for more in a few minutes. Thank you.

Tyler Philpot, Thunder

search

Scott Hebner

>> Hello everyone, welcome back to theCUBE's continuing coverage of Dreamforce 2026. And I'm Scott Hebner, I'm the principal analyst for enterprise AI at theCUBE Research and SiliconANGLE. Thanks so much for tuning in. We continue our conversations here at Dreamforce, and I'm excited to introduce Tyler Philpot, who is the Vice President of Solution Architecture at Thunder. So appreciate you taking the time to be here.

Scott Hebner

>> Thanks for having me.

Scott Hebner

>> Yeah, so why don't we start for our audience, just explain a little bit of what Thunder does and what your specific remit is around solution architecture.

Scott Hebner

>> Thunder's a Salesforce partner. We specialize in implementations and managed services across a variety of different clouds. And my role specifically as VP of Solution Architecture is that I do a lot of the pre-sales, scoping, estimating. Really making sure that our customers' journey is taken into account during the sales cycle so that we can make sure that we're targeting the program and make sure that we're delivering the desired business results and ultimately setting up projects for not just success but continued success time and time again.

Scott Hebner

>> So what do you think of the show so far, in particular Marc Benioff's keynote?

Scott Hebner

>> So far so good. The keynote was great. My favorite part was the walking and talking and the banter between the speakers there.

Scott Hebner

>> And it wasn't lost on me that he was wearing a suit, which we had lost that in the IT industry for decades.

Scott Hebner

>> Oh yeah, he's bringing it back.

Scott Hebner

>> He's always out on the leading edge of doing things, right?

Scott Hebner

>> Right, right, for sure. He is a leader.

Scott Hebner

>> Yeah. So I also noticed that you guys are part of the advisory board for Salesforce as a partner. So that must mean a lot, right, in terms of your authority and how they view you?

Scott Hebner

>> Expertise with the platform, connections, access to a variety of different partner tools and resources. So it really helps us command the marketplace and make sure that we're doing right by our customers.

Scott Hebner

>> Right. And you were one of the early vendors involved with Agentforce,

Scott Hebner

>> right?That's right. Yep. We've got several forward-deployed engineers and that's a product that keeps on running hot. So we're happy to keep going.

Scott Hebner

>> Yeah. Now they're adding AIforce. What's your thought on that? That next layer on the

Scott Hebner

>> architecture?It's the natural next step. Everything's moving in that direction. So we're doing everything we need to do to make sure we're on the cutting edge.

Scott Hebner

>> So a lot of what you do is help connect the pre-sales information to the actual delivery.

Scott Hebner

>> That's right.

Scott Hebner

>> And help keep things in context, right? And that breaking point, what has traditionally happened that's been a problem?

Scott Hebner

>> I'd like to say that's a process that's flawless and there are no gaps in it at all. But we all know that's not true. So even happy path when we have meticulous documentation and we have an accounting of everything a prospect has discussed during the sales cycle, even if we were to transition that seamlessly to the delivery team providing the project, there's always the chance that the client themselves, they have mismanaged expectations internally as well. So that's one thing that we've seen material benefit from working with Certinia and the Veda Accelerate product is not only does it keep us accountable to our own notes, but it's also a great way to make sure that we have strict accounting of who said what in which meetings and what's in scope, what's not in scope, what's been deferred, because the executive sponsor and individual stakeholders a lot of times have different perspectives, and having that complete knowledge graph that's really painting a rich tapestry of who's managing what expectations is really key for us.

Scott Hebner

>> Right, so it's traditionally been a hard problem to solve, because you're dealing with human beings.

Scott Hebner

>> Exactly.

Scott Hebner

>> Now you have AI.

Scott Hebner

>> The human element is always a wild card, so AI is keeping us accountable and keeping everybody on the same page. And that's, you

Scott Hebner

>> know—As I mentioned before we got into our conversation here, I've been a huge proponent of the knowledge graphs and context spaces and things like that. But before we jump into that, tell me a little bit about your architecture, your stack, if you will, that you've built.

Scott Hebner

>> Yeah, well, we are longtime customers of Certinia. So that is our PSA of choice. That's where we manage our projects start to finish. Obviously, Sales Cloud and everything in between there. Slack is a key element of creating that engagement layer where all this information is surfaced and actioned. And then Veda is the new toy to play with. And that's really adding that additional AI layer to make sure that our PSA and our project managers and engagement managers, everyone's working from the same dataset and everyone is looking at proactive signals rather than just reactive responses.

Scott Hebner

>> And that's across, as you said, from pre-sales, absolutely.

Scott Hebner

>> We're able to manage the health and signals from opportunities and the sales cycle, all the way down to, hey, we're coming to UAT next week, what aren't we thinking about that we should be?

Scott Hebner

>> So, AI historically has been very probabilistic in nature. I think Marc Benioff mentioned that they're driving towards more deterministic, but I think a step on that journey is the notion of knowledge graphs.

Scott Hebner

>> Yes.

Scott Hebner

>> And relationships between entities and things of that nature. And you talk a lot about continuous knowledge graphs. Can you just give our audience a little bit of a sort of an overview of what you mean by that and then what it sort of does and how you're doing it?

Scott Hebner

>> Yeah, when it comes to knowledge graph, it really, I've heard it said that, prompt engineering was the term of art, but now it's really context engineering. That's really what matters most. And I can't agree with that more because, when we're thinking about just how we balance continuity from a presale cycle into a project with varying perspectives and inputs and outputs and decisions being made along the way. That knowledge graph is continuously being changed. So what was scoped, in presales may have changed 4 times by the time you get to UAT. So you need to make sure you're continuously updating and refining what that knowledge graph looks like and really take into consideration everything, all the discussions and inputs that it's been able to analyze throughout the project.

Scott Hebner

>> Is the context part of what comes from that knowledge graph? Or is that an additional layer, like what they do with Moonnox?

Scott Hebner

>> Yeah, the context, it's coming from a variety of sources. We have our Google Drive, we have our PSA data, we have Slack channels. All of this is creating this rich tapestry of context. It really comes down to which context supersedes the other. How do we avoid overfitting context to make sure that the knowledge graph is complete, accurate, but also void of superfluous noise, for lack of a better term.

Scott Hebner

>> So when you're talking to a client or a potential prospect, do you get into discussing the knowledge graph and the power of it? Is it resonating?

Scott Hebner

>> It is a selling factor for sure. Just knowing that we're AI-enabled and we have these tools at our disposal to make sure that we're able to run their project. More successfully and ultimately translate into savings for them, I think that means a lot. And knowing that they're not going to have to repeat themselves also, because I think that's a common complaint across clients, that, hey, we're doing discovery, we've got, 3 weeks of in-depth conversations with all these individual stakeholders, with the architects and, the engagement managers running the project. But you know what, I told your pre-sales guy that, a month ago. Why do I have to repeat myself? So a lot of this is able to, or enables us to be able to build on that knowledge and add to it rather than just rehashing it across meetings.

Scott Hebner

>> So in many ways, it's building institutional knowledge about the project, about the client.

Scott Hebner

>> Absolutely.

Scott Hebner

>> Yeah. It takes the more general purpose normalized data that you have from all your engagements and starts to build a knowledge graph that applies that to that specific project. That's basically the value of it, where in the past it was more LLM-based, probabilistic. Yeah. You weren't building institutional knowledge.

Scott Hebner

>> Right, right.

Scott Hebner

>> And then increasingly having context around that.

Scott Hebner

>> Absolutely. Yeah, it's been huge. So that speeds the delivery cycle, lowers costs, and provides a better customer experience as well. And that's one of those intangibles that I think really goes a long way in making sure that the consulting relationship is as smooth and seamless as possible.

Scott Hebner

>> And one of the themes here from Certinia is Be Certain, right? And the system of action and you need to be increasingly certain about the actions you take and how they affect the outcomes. And I would imagine there's a direct correlation or tie to the knowledge graph and the context, right? Because you can't make a good judgment and Be Certain about it and feel good about it unless you understand the context.

Scott Hebner

>> Just like analytics, garbage in, garbage out. You got to make sure that the knowledge is clean and specific and tangible and helps inform those AI agents to be able to make those recommendations based on good information.

Scott Hebner

>> And one thing I learned about you guys is a term that I haven't really heard out there, but I think it's super intriguing, which is a context engineer. Why don't you just talk? I think I got the idea here, but for the, you know, I'm assuming it's someone that, like you mentioned prompt engineering. Yeah, is this a new discipline, evolved discipline, or?

Scott Hebner

>> I think it's quickly becoming, if it's not there already, the newest term of art that people are really focusing on. It's all about establishing your pipeline to make sure that by the time you get to your agent and you start prompting it to do something, you want to make sure that by that point it has the relevant context. And I think that's one thing that we're challenged with is how do we provide it with the documents that we got from our partners, with the, you know, interviews and the discovery sessions we had with our prospects and all these other sources of input. How do we curate it in such a way that by the time I get to that agent, it knows, hey, these are the documents that are relevant. Here's the context that I should be drawing from when responding to that prompt. Too broad context, you get more hallucinations and more difficult answers. But if you get very specific and very precise with your context, you get better answers.

Scott Hebner

>> So as you lead the solution architecture for Thunder, you have context engineering activities underway? Absolutely. And that's an interesting approach because I have not heard that term before, but I think it's important because it is different. You're right, it is different than prompt engineering. That's very different. And it kind of takes this notion that generative AI, the LLMs, are sort of like the browsers of the early— it's the gateway into the world of AI. But to really be successful, you got to build institutional knowledge around your particular domain business.

Scott Hebner

>> For sure. Yeah.

Scott Hebner

>> And that's the knowledge graph, and you have the context coming from it, and you need the discipline to be able to build and maintain all that. And I think that's what you're getting at.

Scott Hebner

>> Absolutely. Yeah.

Scott Hebner

>> Interesting. So where do you think this is all heading? Just in following professional services pretty much this full year, and Certinia in particular, there's been a lot of innovation. I just think back to March, and then they got into agentic knowledge. They were talking about knowledge graphs, and they had the Moonnox, and now the record system of— not record, but System of Action. What do you think in 3 months we're going to be talking about? Today is sort of like knowledge graphs and context, do you think it's moving quick enough that we have the next big horizon here for what you guys do?

Scott Hebner

>> Yeah, some of the big advances as of late, we have MCP servers becoming more ubiquitous and more autonomous agents working, and then we have, you know, headless architectures where we're no longer bound to these UIs. So it's only a matter of time, I think, until you have projects being run semi-autonomously through these headless solutions. It's going to be a wild ride, that's for sure. I'm certain of that.

Scott Hebner

>> Yeah, well, there's been a theme in some of the interviews I've been doing about people making comments that AI is actually generating more jobs and more work, which is a good thing because you're able to do your projects quicker, more efficiently, lower cost. Which then incents the client to do more. So it's not actually eating away people's jobs in professional services. Is that your observation too?

Scott Hebner

>> I haven't seen it yet. It's been able to take a lot of the administrivia, the low-level work, and allow that higher-level strategic thinking to take place. And we do want to make sure that at least at Thunder, we're not getting into that role of just ticket taker develop the solution. But let's make sure that, hey, if we can send off the digital labor to build the solution. Let's spend our extra time making sure that we're building the right solution for the right outcomes and really making sure that everything is buttoned up that way. And I haven't— yeah, I lost my train of thought there. Sorry.

Scott Hebner

>> Yeah, no, I think there's just a lot of— you look on the broadcast news and all the fear about AI taking everyone's jobs, and I just don't— I personally, I was a CMO for IBM's AI business for a while before I retired, then I've been an analyst ever since following this, and I just don't see the massive reduction of jobs. If anything, they're shifting, to your point.

Scott Hebner

>> There's definitely shifting. I think the roles will continue to be there, so we're still going to need architects, developers, analysts on projects, but the nature and how they work with their tools and with their digital labor counterparts, I think that'll be exciting to see. Faster throughput, maybe deployed to more projects. So there's a lot of things that can come out of it.

Scott Hebner

>> Yeah, I think people should be excited about it. And if you're in professional services, it's giving you superpowers to be more valuable and to do more things and have a better next step of career moves because you learn how to use AI effectively, you're going to be a lot more valuable than someone that doesn't.

Scott Hebner

>> It's definitely going to be an impactful technology to learn. It's either you adopt it and you become a multiplier, or if you lag behind, options would be a little more limited, I would imagine.

Scott Hebner

>> Yeah, I think the context engineering is a good example of that, sort of a new discipline, because you still need prompt engineers. Yeah, right. So it's not like, they get replaced. Now, some of those might be dual jobs, but again, it's creating more and more opportunities. one thing I believe in, the reason why I cover professional services, I think it is, if you want to know what's going to happen in enterprise AI, you look at professional services as an industry, because you guys by definition are ahead of the game, because just by the nature of what you do, you build solutions. Right, right. And how AI affects you and what you're able to do and the value you get out of it, it's ultimately going to happen in the other industries. Yep. So what happens in professional services is going to then happen. That's why I've been watching you guys so carefully. And you guys have made the move as an industry from a technology discussion to economics and outcomes, which I think is really healthy. Yeah.

Scott Hebner

>> Yep. That's the trend for sure.

Scott Hebner

>> A lot of the other industries I haven't seen that yet. Which brings me to the last question, just to get your two cents on, is this move potentially to fixed-price bids, outcome-based pricing versus historically obviously not that way. It was based on billable hours and materials.

Scott Hebner

>> Yeah.what do you see happening now?

Scott Hebner

>> Ultimately, we want to make sure that we're driving consistency and predictable outcomes with our customers, but also, financial side of things. If you get twice as efficient and you're working on a time and materials basis, it means you get half as much revenue. So that's the mathematics of the situation. So part of this is definitely customer base and making sure that we're driving predictable results. But also how do we protect our margins at the same time and make sure that digital labor isn't cutting into our revenue?

Scott Hebner

>> Right.

Scott Hebner

>> Yeah.No, I think it goes back to that notion that you do it quicker, more efficiently, at a lower cost, you delight a customer, then they're going to want to do more. And there seems to be an endless appetite for the other thing.

Scott Hebner

>> Exactly.And as AI keeps speeding along in terms of innovation, there's only going to be new things to do. So it seems like a very healthy dynamic that's occurring. Yeah. We'll come and take down into the other industries, just sort of leading the charge.

Scott Hebner

>> So success begets more success. So we're excited to use our AI and delight our customers and make sure that they're on a long journey with

Scott Hebner

>> us.You get to be a guinea pig.

Scott Hebner

>> Exactly.

Scott Hebner

>> Yeah.

Scott Hebner

>> Yeah.It's fun. No, no dull

Scott Hebner

>> moments.All right. Great conversation.

Scott Hebner

>> Yeah.Appreciate you being here. Appreciate

Scott Hebner

>> that.Thank you. All right. Well, thank you all for taking the time to be here. We will continue our coverage of Dreamforce 2026. Stay tuned for more in a few minutes. Thank you.