Taking on ServiceNow: Coding Agents the Rest of Us with Serval Co-Founder and CEO Jake Stauch

“This is the ultimate moonshot in enterprise software. These agents that go off and do work for you and magically discover all your problems and solve all your problems. And I think no one's really figured out how to do this well.”

Claim: The speaker views enterprise agents that autonomously discover and solve users’ problems as the ultimate moonshot in enterprise software and says no one has yet figured out how to build them well.

“The agent becomes pretty competent. It could theoretically just start writing code that could take action on all your systems, and that is an unbounded security risk that I don't think anyone's gonna be able to take anytime soon.”

Claim: The speaker believes that a sufficiently competent agent could theoretically write code to act across all of a user's systems, creating an unbounded security risk that no one will be able to take on anytime soon.

“On the one hand, you want these to be very autonomous without somebody babysitting them. On the other hand, autonomy is very scary in the enterprise. Serval is the boundary. Serval is the guardrails.”

Claim: The speaker describes Serval as the boundary and guardrails that constrain enterprise agents.

“On the one hand, you want these to be very autonomous without somebody babysitting them. On the other hand, autonomy is very scary in the enterprise. Serval is the boundary. Serval is the guardrails.”

Claim: The speaker says enterprises want agents to operate autonomously without human babysitting, while also finding enterprise autonomy very concerning.

“We are the way that a large enterprise with a lot of rules and approvals and process and politics can adopt these tools safely and comfortably at scale.”

Claim: The speaker says Serval enables large enterprises with extensive rules, approvals, processes, and internal politics to adopt AI tools safely and comfortably at scale.

“Yeah, so background, Serval is enterprise automation, enterprise service management. The idea is you can ask questions and get help at work and automate a lot of the parts of the job you don't enjoy, and we do that through code gen agents.”

Claim: Serval provides enterprise automation and service management through code-generation agents that answer workplace questions, provide help, and automate tasks employees dislike.

“So that's the-- what the company does. We compete with companies like ServiceNow, is historically the big legacy provider in the space. I discovered this category actually working at a previous company called Verkada, uh, where we sold enterprise physical security products to IT teams.”

Claim: Serval competes with ServiceNow, which the speaker describes as the historically dominant legacy provider in enterprise service management.

“So that's the-- what the company does. We compete with companies like ServiceNow, is historically the big legacy provider in the space. I discovered this category actually working at a previous company called Verkada, uh, where we sold enterprise physical security products to IT teams.”

Claim: The speaker says he discovered enterprise service management while working at Verkada, a company that sold enterprise physical security products to IT teams.

“Spent many, many years talking to IT teams, figuring out what to build for them, and one of the things I love to, to do with these teams is ask them, "What do you spend the most money on, and what do you hate the most?"”

Claim: The speaker spent many years consulting IT teams about their needs and what products to build for them.

“And ServiceNow would top a lot of those lists. And so I was always interested in this category of maybe there's a better solution here, but it wasn't clear for most of my time there how you'd ever beat ServiceNow.”

Claim: The speaker says that for most of his time at Verkada, he could not see how a competitor could beat ServiceNow.

“And ServiceNow would top a lot of those lists. And so I was always interested in this category of maybe there's a better solution here, but it wasn't clear for most of my time there how you'd ever beat ServiceNow.”

Claim: The speaker says ServiceNow frequently ranked among the products IT teams spent the most money on and disliked the most.

“They're a great company. They've done so much so well, and why would you be able to compete with them? And I-”

Claim: The speaker considers ServiceNow a strong company that has executed many things well.

“I think so hard to build in the platform, so it's a very robust, customizable, powerful platform that can map to all these enterprise workflows.”

Claim: The speaker says ServiceNow is a robust, customizable platform capable of mapping enterprise workflows, but building on it is very difficult.

“Fundamentally, it's a workflow automation platform, and so it can do so much, but every idea you have for doing something cool, you have to go and implement, and a lot of times you're bringing in consultants to implement it.”

Claim: The speaker describes ServiceNow as a workflow automation platform where new capabilities must be implemented, often by bringing in consultants.

“So it, it's always kinda running six to twelve months behind where you want it to be. And it's fine if you're building something that's gonna last five, ten years, like a big onboarding flow, and nothing about your business is gonna change, so you can invest all that upfront effort, and then it gets deployed, and it works for a while.”

Claim: The speaker says ServiceNow implementations commonly lag users’ desired state by six to twelve months.

“So it, it's always kinda running six to twelve months behind where you want it to be. And it's fine if you're building something that's gonna last five, ten years, like a big onboarding flow, and nothing about your business is gonna change, so you can invest all that upfront effort, and then it gets deployed, and it works for a while.”

Claim: The speaker says ServiceNow can suit long-lived projects, such as a major onboarding flow, when business needs are stable enough to justify substantial upfront implementation effort.

“But for just trying to keep up with all the automation that you wanna build, ServiceNow is not able to catch up. And, and then it's very expensive, and it keeps getting more expensive over time, so that frustration of it's not keeping up with where I wanna go as a business, and it keeps getting more expensive, that's what was driving a lot of the frustration.”

Claim: The speaker says ServiceNow cannot keep pace with businesses’ desired automation and is very expensive, with costs continuing to rise over time.

“Yeah, we love this idea of coding agents for the rest of us because if you think, so much of the early enterprise adoption in AI has been engineering teams. And if you think about why that is, there's a lot of reasons, obviously, early adopters in a lot of different technologies, but also they're able to build and experiment with these tools because they have this process of code review.”

Claim: The speaker says engineering teams account for much of early enterprise AI adoption, in part because code review provides a process for safely building and experimenting with AI tools.

“It's very safe for even a nineteen-year-old intern to build something very powerful and then go through this process before it goes into production. We wanna bring that same capability to the rest of the enterprise, so we build a platform that brings coding agents to IT departments who then can bring it to HR and finance and people teams and legal teams.”

Claim: The speaker says code review can allow even a nineteen-year-old intern to build a powerful tool that is checked before production deployment.

“It's very safe for even a nineteen-year-old intern to build something very powerful and then go through this process before it goes into production. We wanna bring that same capability to the rest of the enterprise, so we build a platform that brings coding agents to IT departments who then can bring it to HR and finance and people teams and legal teams.”

Claim: The speaker says Serval brings coding agents to IT departments so they can make the capabilities available to HR, finance, people, and legal teams.

“And they can do it because under the hood, it's coding agents, but then you have all these abstractions and guardrails and controls, so things go through approval. They go through a staging and, uh, approval cycle and audit trail and all of that.”

Claim: The speaker says Serval wraps coding agents in abstractions, guardrails, and controls, including staging, approval cycles, and audit trails.

“And so it's how do you build a system that makes it easy for everyone to be a builder? And the core insight here was that IT actually wants to build stuff. And for most of the world, they don't have an engineering department.”

Claim: The speaker says IT departments want to build things and that most organizations lack a dedicated engineering department.

“They have an IT department. Most companies have an IT department, not, not a, a team of software engineers. And so IT is what's building and configuring technology for the company, and we're building essentially the AI platform for them.”

Claim: The speaker says Serval is building an AI platform for IT departments.

“They have an IT department. Most companies have an IT department, not, not a, a team of software engineers. And so IT is what's building and configuring technology for the company, and we're building essentially the AI platform for them.”

Claim: The speaker says most companies have IT departments rather than software engineering teams, and that IT is responsible for building and configuring company technology.

“Yeah, I think those really work well for consumers and SMB and even a lot of the mid-market organizations. In a large enterprise, there's so much, and I didn't even really appreciate this going in, there is so much process around approvals, around what you can and can't do.”

Claim: Workforce agents and virtual assistants work well for consumers, small and medium-sized businesses, and many mid-market organizations.

“Yeah, I think those really work well for consumers and SMB and even a lot of the mid-market organizations. In a large enterprise, there's so much, and I didn't even really appreciate this going in, there is so much process around approvals, around what you can and can't do.”

Claim: Large enterprises have extensive processes governing approvals and what employees can and cannot do.

“The permission sets of any individual are so limited that even if you are allowed to bring your agent to work, which a lot of cases you're not, what it can, what it can do is so incredibly limited.”

Claim: In many cases, employees are not allowed to bring their own AI agents to work.

“The permission sets of any individual are so limited that even if you are allowed to bring your agent to work, which a lot of cases you're not, what it can, what it can do is so incredibly limited.”

Claim: Even when employees are allowed to bring AI agents to work, their limited permissions constrain what those agents can do.

“It can, it can barely even read your emails. Um, and so when you think about all the stuff you need to do at work and all the systems you wanna touch and all the interesting automations you could build, your personal permissions end up being so limited that these personal agents are also incredibly limited and all-- not all that useful.”

Claim: Because employees' permissions are so limited, personal AI agents can barely read their emails and are not very useful for many workplace tasks and automations.

“So what we can do, because we work with these IT teams that can connect with, you know, service account credentials and admin-level permissions, we can create a structure where people can access elevated permissions and do all these crazy things in the org, but with very, very strict deterministic approval cycles, workflows that are deterministic, um, and an audit trail because under the hood, we are also a ticketing system.”

Claim: Serval works with IT teams that can connect through service-account credentials and admin-level permissions.

“So what we can do, because we work with these IT teams that can connect with, you know, service account credentials and admin-level permissions, we can create a structure where people can access elevated permissions and do all these crazy things in the org, but with very, very strict deterministic approval cycles, workflows that are deterministic, um, and an audit trail because under the hood, we are also a ticketing system.”

Claim: Serval provides access to elevated permissions through deterministic approval workflows and an audit trail, and its underlying product also functions as a ticketing system.

“So you say, "Okay, I want my agent to go do X, Y, and Z." Well Most organizations are gonna be very uncomfortable with that at the large enterprise scale but with Serval, that's gonna create access requests that go through an approval process, and all of that's gonna be documented, logged, and then you can get just-in-time access to be able to run that deterministic workflow one time, and all of it's gonna be tracked.”

Claim: In large enterprises, Serval routes agent task requests through documented and logged approvals, then grants tracked just-in-time access for one execution of a deterministic workflow.

“And yes, it's a little bit more annoying than the magical personal agent experience we love with Instinct and Muse and all these tools, but it's what has to be true for a large enterprise, especially a regulated enterprise, to ever adopt these kinds of technologies.”

Claim: Although Serval's approval-based workflows are less convenient than personal agents, they are necessary for large and especially regulated enterprises to adopt agent technologies.

“Yeah, absolutely. Diffusion is obviously the gap here. The technology can do so much and, and enterprises have barely scratched the surface of what's possible, and a lot of that is just understanding the people, the process, what workflows actually exist, how you want them to exist in the future, and we think that you still need people for that.”

Claim: Enterprises have barely begun to realize the potential of AI technology.

“Yeah, absolutely. Diffusion is obviously the gap here. The technology can do so much and, and enterprises have barely scratched the surface of what's possible, and a lot of that is just understanding the people, the process, what workflows actually exist, how you want them to exist in the future, and we think that you still need people for that.”

Claim: Enterprise adoption of AI requires understanding people and existing and desired workflows, so human involvement is still needed for that work.

“So we deploy both a, a deployment strategist we call them, and a f- deployed software engineer that go on site with customers, figure out how do we implement this?”

Claim: Serval sends deployment strategists and forward-deployed software engineers to customer sites to plan implementation.

“How do we solve all your problems as a business? And then walk you through the implementation. What we found, though, is that a lot of times after we've got this implemented and deployed, customers then have a long tail of more stuff they want.”

Claim: Serval customers often request substantial additional work after initial implementation and deployment.

“And, and as we've discovered with software engineering, there's actually unlimited demand for work. And we've started to build out a team, we call it Pro Serval, uh, or, or automation engineering is, is the field.”

Claim: Serval's experience with software engineering suggests that demand for work is effectively unlimited.

“And, and as we've discovered with software engineering, there's actually unlimited demand for work. And we've started to build out a team, we call it Pro Serval, uh, or, or automation engineering is, is the field. And we bring in these automation engineers to kind of do ongoing continued support and services because there's so much that they want to automate, and they need the help to be able to figure out what can be automated?”

Claim: Serval has built an automation-engineering team called Pro Serval to provide ongoing customer support and services.

“And we bring in these automation engineers to kind of do ongoing continued support and services because there's so much that they want to automate, and they need the help to be able to figure out what can be automated? What do the processes look like? How do we navigate all of our systems and then put it into production with a product like Serval?”

Claim: Serval customers want to automate many tasks and need help identifying automatable work, understanding processes and systems, and deploying automations into production.

“I like to say everything has gone exactly according to our very delusional plans. So we set out from the beginning to be the largest enterprise software company of all time. We actually made a list of all the companies.”

Claim: Serval's founding ambition was to become the largest enterprise software company of all time.

“We circled ServiceNow, and we said, "We're gonna beat them, and we're gonna build something much bigger." We said that on day one when it was two people in my co-founder Alex's living room, and we just said, "We're gonna do it."”

Claim: From its founding, Serval intended to surpass ServiceNow and build a substantially larger company.

“And so even though that was crazy, that was always the intention. We didn't grow into that ambition. We set out from the beginning. I think that that actually-”

Claim: Serval began with the ambition to become larger than ServiceNow; that ambition did not develop only after the company grew.

“So the, the way you typically go after these, these large enterprise startups at least, they, they build like the layer- ... or they go down market, and those are kind of the two common approaches. Let's build a small layer, play nice with the big incumbents, AI agent on top of Salesforce-”

Claim: One common strategy for large enterprise startups is to build a narrow product layer that complements incumbents, such as an AI agent on top of Salesforce.

“... which like it, you know, it, it really caps how big you can get.”

Claim: A startup that builds a small layer alongside a large incumbent faces a ceiling on how large it can become.

“It's really tough. It like caps you unless you find some niche to kind of like, uh, expand into later on. Um, the other approach is we just go downmark- you know, go downmarket, service the SMB and mid-market, and you see that with a lot of, you know, competitors to Workday, competitors to Salesforce.”

Claim: A startup building a layer alongside an incumbent is difficult to scale unless it later finds a niche for expansion.

“It's really tough. It like caps you unless you find some niche to kind of like, uh, expand into later on. Um, the other approach is we just go downmark- you know, go downmarket, service the SMB and mid-market, and you see that with a lot of, you know, competitors to Workday, competitors to Salesforce. They just say, "We're gonna be that but for the lower end." And we just looked at-”

Claim: Some enterprise software startups pursue the small and mid-market by offering lower-end alternatives to products such as Workday and Salesforce.

“Exactly. And we looked at that, and we said, "Why not just go after the big one?" It's, it's much harder, but at least the prize is quite big if you win. And I think that was so important for us because you end up making so many different decisions if you're building for the large enterprise from day one.”

Claim: Building for large enterprises from day one is harder than starting with one market and expanding later, but offers a larger potential payoff if successful.

“All the product decisions, how you build your team, how you think about growth, they're fundamentally different than if you go and you serve one market and then you hope to expand later on.”

Claim: Building for large enterprises from day one leads to fundamentally different product, team, and growth decisions than starting with one market and expanding later.

“I think we didn't really understand how we'd beat the other companies and why we had an advantage and what, what AI unlocked that they couldn't catch up to. And what we understood about ServiceNow because of our experience with IT is that fundamentally it is a workflow automation platform.”

Claim: ServiceNow is fundamentally a workflow automation platform.

“I think we didn't really understand how we'd beat the other companies and why we had an advantage and what, what AI unlocked that they couldn't catch up to. And what we understood about ServiceNow because of our experience with IT is that fundamentally it is a workflow automation platform.”

Claim: The founders initially did not understand how they could beat other companies, what advantage they had, or what AI enabled them to do that competitors could not catch up to.

“And our belief, me and my co-founder, and my co-founder really had this, this idea early on that code was the right way to express automation in the enterprise, that you should just build all automations in code, and these drag-and-drop workflow builders is a painful level of abstraction because it's too complex for non-technical users, and it's annoyingly abstracted for technical users.”

Claim: Jake Stauch and his co-founder believed code was the right way to express automation in the enterprise.

“And our belief, me and my co-founder, and my co-founder really had this, this idea early on that code was the right way to express automation in the enterprise, that you should just build all automations in code, and these drag-and-drop workflow builders is a painful level of abstraction because it's too complex for non-technical users, and it's annoyingly abstracted for technical users.”

Claim: Drag-and-drop workflow builders are a poor abstraction: they are too complex for nontechnical users and too abstract for technical users.

“Code is a better way of building automations. The problem obviously is not many people can write code and, and do it quickly enough for it to make sense for automation. And so as AI made this possible, that was what it clicked for both of us that actually if code is the right way to build automation and now anyone can write code, we can build a better automation platform.”

Claim: The founders saw AI as making code-writing accessible to anyone, creating an opportunity to build a better automation platform.

“Code is a better way of building automations. The problem obviously is not many people can write code and, and do it quickly enough for it to make sense for automation. And so as AI made this possible, that was what it clicked for both of us that actually if code is the right way to build automation and now anyone can write code, we can build a better automation platform.”

Claim: Few people can write code quickly enough for code-based automation to be practical without AI assistance.

“Code is a better way of building automations. The problem obviously is not many people can write code and, and do it quickly enough for it to make sense for automation. And so as AI made this possible, that was what it clicked for both of us that actually if code is the right way to build automation and now anyone can write code, we can build a better automation platform.”

Claim: Code is a better way to build automations than drag-and-drop workflow builders.

“And what is an automation platform is actually b- what is the best market for an automation platform? It's actually taken on these massive legacy software companies like ServiceNow.”

Claim: Stauch believes a strong market for an automation platform is competing with massive legacy software companies such as ServiceNow.

“The bet was that coding agents would be-- would get very good. Um, and the, the theory there was it's a verifiable domain, and, and it should actually hill climb fairly well.”

Claim: The founders bet that coding agents would become very capable because coding is verifiable and should support steady iterative improvement.

“And so we believed fundamentally that coding agents would get good, at least for these very... I didn't expect them to get as good as they are for just general purpose software development but for deterministic procedural scripts of reset password, get approval, like all of these things are just so defined and verifiable that we felt like it would get good at those things.”

Claim: Stauch says he did not expect coding agents to become as capable at general-purpose software development as they have become.

“And so we believed fundamentally that coding agents would get good, at least for these very... I didn't expect them to get as good as they are for just general purpose software development but for deterministic procedural scripts of reset password, get approval, like all of these things are just so defined and verifiable that we felt like it would get good at those things.”

Claim: The founders expected coding agents to become capable at deterministic, procedural enterprise tasks such as password resets and approvals because those tasks are well-defined and verifiable.

“And so from a workflow automation perspective, we thought it would get there. In the early days, it did not. So what we ended up doing in all our early demos, we would prompt the LLM and say, "Hey, here's a-" ... suggested way you might wanna build a password reset workflow.”

Claim: In the early days, coding agents were not good enough for workflow automation.

“And so from a workflow automation perspective, we thought it would get there. In the early days, it did not. So what we ended up doing in all our early demos, we would prompt the LLM and say, "Hey, here's a-" ... suggested way you might wanna build a password reset workflow.”

Claim: In early demos, the team used an LLM prompt to suggest how to build a password-reset workflow.

“And so we'd, we'd, we'd seed it with, like, a lot of examples, and then over time we'd make that library bigger and bigger. But then the models just got so good that eventually it became hard to break.”

Claim: The models improved to the point that they eventually became difficult to break in the team's workflow-automation use cases.

“And so we'd, we'd, we'd seed it with, like, a lot of examples, and then over time we'd make that library bigger and bigger. But then the models just got so good that eventually it became hard to break.”

Claim: The team initially supplied many examples to its system and enlarged the example library over time.

“It became hard for you to give it a workflow that it couldn't automate if the APIs existed.”

Claim: Once the necessary APIs existed, the system could automate nearly any workflow the team tried to give it.

“We're so excited for them to get better because I think it unlocks a lot of the things that we'd love to do around, you know, better browser use and computer use and more of a approach to background agents and personal agents that can be really useful in the enterprise.”

Claim: Stauch expects better coding agents to enable improved browser and computer use, as well as useful enterprise background and personal agents, and is excited about that prospect.

“So we're excited for the evolution of the category. We've always believed fundamentally that we are not building the LLMs ourselves, um, we are building the guardrails.”

Claim: Serval does not build its own LLMs; it builds guardrails around them.

“We are building, um, the, the boundaries, and the boundaries are the product for us because we are the way that a large enterprise, uh, with a lot of rules and approvals and process and politics can adopt these tools safely and comfortably at scale.”

Claim: Serval's product is the boundaries and guardrails that help large enterprises adopt AI tools safely and comfortably at scale despite their rules, approvals, processes, and internal politics.

“And so Serval is the boundary. Serval is the guardrails. The intelligence getting better just makes our product more powerful and also makes our customers wanna actually clamp down even further on those boundaries, on those guardrails.”

Claim: Improving AI intelligence makes Serval's product more powerful and leads its customers to want tighter boundaries and guardrails.

“And I actually think access is going to get more and more controlled for individuals in the enterprise. And as access becomes stricter and you let people do less and less and your permissions get m- more and more narrow, which I think is going to happen-”

Claim: Stauch predicts that enterprises will increasingly control individual access and narrow employees' permissions.

“Because I think if, if you think about, you know, I might safely give you more permissions than you need because I don't actually think you're gonna do all that much damage.”

Claim: An organization may grant a person more permissions than necessary when it believes that person is unlikely to cause significant damage.

“But if you have access to all these agents, then the damage you can do with any kind of, like, slightly elevated beyond normal permissions expands, and, and we're seeing this in cybersecurity risk right now.”

Claim: Access to many agents increases the potential damage from permissions that are only slightly above normal, a pattern Stauch says is already visible in cybersecurity risk.

“And, and so what I'm actually gonna wanna do if I'm a large enterprise is shrink everyone's permissions down to truly least privilege. Like, you cannot access any document, and you cannot take any action unless you absolutely need to do it for your job.”

Claim: Stauch expects large enterprises to reduce employees' permissions to least privilege, allowing document access and actions only when required for their jobs.

“But then you're going to have exceptions all the time. We are the layer that grants those exceptions and makes it possible for you to set up approval procedures around that, have an audit trail and a ticket associated with that.”

Claim: Enterprises will continually need exceptions to least-privilege access, and Serval provides a layer for granting them through approval procedures with audit trails and associated tickets.

“So by owning how privileges get escalated in the enterprise, we actually become more powerful and more essential the more you clamp down on end user permissions.”

Claim: Stauch argues that Serval becomes more powerful and essential as enterprises tighten end-user permissions, because Serval controls privilege escalation.

“I think we have to convince ourselves and our customers that Serval is the way you get promoted at work. It, it can't be the fact that Serval is a risk that you take to hopefully get something on the other side.”

Claim: Jake Stauch says Serval must be presented to customers as a way to earn a promotion, rather than as a risky bet with an uncertain payoff.

“It's a investment of their own time, I assume. It's, like, personal risk. Like, why do that? Or is there a, a... Is there an experience where people have already gone through to the other side?”

Claim: Jake Stauch says Serval should help customers advance their careers and earn promotions, not merely avoid negative consequences.

“Top-down. So we sell into the C-level- ... sell into IT leadership- ... in, in many cases, and then distribute into other departments. And that is the message is this is people built their career on ServiceNow. You're gonna build your career on this next generation.”

Claim: Jake Stauch says Serval sells top-down to C-level executives and IT leadership, then often expands into other departments.

“And we're seeing this play out. So one of our largest customers is Fox. Um, Fox Corporation adopted Serval company-wide, and the person who led that migration at, on the g- on the ground level actually got nominated to be the MVP of the entire company.”

Claim: Jake Stauch says Fox Corporation adopted Serval company-wide and nominated the employee who led the migration on the ground to be the company's MVP.

“And, and got an award for his work there, and that is really what we're about, is we're convincing our customers, saying, like, "This is the way to not just prove that you're, you're useful to your organization, but to prove that you might be the most essential person in your company because you're leading this change."”

Claim: Jake Stauch says the employee who led Fox Corporation's Serval migration received an award for his work.

“And, and got an award for his work there, and that is really what we're about, is we're convincing our customers, saying, like, "This is the way to not just prove that you're, you're useful to your organization, but to prove that you might be the most essential person in your company because you're leading this change."”

Claim: Jake Stauch says leading an AI-related organizational change can demonstrate an employee's value and might establish them as the company's most essential person.

“And we're so pivotable, pivotal to the AI transformation story that I think it's a, it's something we can sell really nicely to our customers.”

Claim: Jake Stauch believes Serval is pivotal to AI transformation and that this positioning can be sold effectively to customers.

“I think consumer often gives us a win- window into what's gonna happen next in enterprise. Um, and it all... it doesn't always look the same, but I think it's a, it's a nice window, so I, I'm very bullish on what do personal agents look like in the enterprise.”

Claim: Jake Stauch believes consumer products often foreshadow enterprise developments, although enterprise outcomes do not always resemble consumer products.

“I think consumer often gives us a win- window into what's gonna happen next in enterprise. Um, and it all... it doesn't always look the same, but I think it's a, it's a nice window, so I, I'm very bullish on what do personal agents look like in the enterprise.”

Claim: Jake Stauch is very optimistic about the future of personal agents in enterprise settings.

“I think in the same way that email and Slack and these tools transformed how we communicate at work, I think personal agents will also be a big part of that.”

Claim: Jake Stauch predicts personal agents will become a major part of workplace communication, analogous to the impact of email and Slack.

“I think everyone will have their personal agent that does tasks for them, that delegates to other personal agents, and so your agent will talk to other agents at work. So I'm really excited about that, but then how do we secure that in a world of increasing cybersecurity risk and increasing concerns at large enterprises around what's happening with everyone's agents taking action in all their computers and, and using their browsers?”

Claim: Jake Stauch predicts that everyone will have a personal agent that performs tasks, delegates to other agents, and communicates with workplace agents.

“I think everyone will have their personal agent that does tasks for them, that delegates to other personal agents, and so your agent will talk to other agents at work. So I'm really excited about that, but then how do we secure that in a world of increasing cybersecurity risk and increasing concerns at large enterprises around what's happening with everyone's agents taking action in all their computers and, and using their browsers?”

Claim: Jake Stauch says securing personal agents will be challenging as cybersecurity risks rise and enterprises worry about agents acting across employees' computers and browsers.

“And so that tension between what employees, consumers want out of their personal agents at work and what enterprises need from a governance perspective, I think is really interesting, and I think there's gonna be a lot of opportunity in that, in that tension of-”

Claim: Jake Stauch believes the conflict between users' desired workplace personal-agent capabilities and enterprises' governance requirements will create substantial opportunities.

“I think so Yeah. I think so. In the same way that I have a Gmail and I have a- ... a, a corporate email. Um, yeah.”

Claim: Jake Stauch expects each person to have separate personal and work agents, analogous to separate personal and corporate email accounts.

“I think the biggest one is the software development life cycle, something that we take for granted in engineering context of the, "Hey, there's actually like a dev environment, and then there's staging, and then there's prod, and there's like these gates between them, and maybe you've got, you know, branches that people can develop on that's not gonna harm the other systems."”

Claim: Jake Stauch says Serval's key design principle is to apply software-development-life-cycle practices to automation, including separate development, staging, and production environments, gates, and isolated branches.

“Uh, it does in some cases, but I think most new AI products are not embracing that. They're basically saying, "Hey, let's build these cool agents and ship to production."”

Claim: Jake Stauch says some IT automation follows software-development-life-cycle practices, but most new AI products do not and instead build agents for direct production deployment.

“But this idea of, hey, you have to actually follow an SDLC process if you're a large enterprise. A large financial institution is not just going to ship new agent capabilities, uh, without a, a review and user acceptance testing and, and all of these-”

Claim: Jake Stauch says large enterprises, including financial institutions, require an SDLC process with review and user acceptance testing before deploying new agent capabilities.

“And so I think that's something we think about a lot is what is going to convince the largest companies in the world to adopt this technology, not just what is powerful and cool and something we're excited to get in the hands of customers.”

Claim: Jake Stauch says Serval focuses on what will persuade the world's largest companies to adopt AI technology, not only on its capabilities or novelty.

“I, I think in some cases they never get there. Like some companies just, they, they look at what enterprise software takes, and they decide, "We don't really wanna build that." And it, it takes... You know, you, you look at the story of a lot of these companies that we mentioned earlier that have gone down market, I think sometimes they do look at the opportunity to go and build for the enterprise, and it, it requires a whole different set of building principles and products that you have to build, and some just don't choose to do it, and they, they cede that ground to other companies.”

Claim: Some companies choose not to build enterprise software because it requires a different set of building principles and products, leaving that market to other companies.

“I, I think in some cases they never get there. Like some companies just, they, they look at what enterprise software takes, and they decide, "We don't really wanna build that." And it, it takes...”

Claim: Jake Stauch says some AI companies decide against building enterprise software after considering what it requires.

“You know, you, you look at the story of a lot of these companies that we mentioned earlier that have gone down market, I think sometimes they do look at the opportunity to go and build for the enterprise, and it, it requires a whole different set of building principles and products that you have to build, and some just don't choose to do it, and they, they cede that ground to other companies.”

Claim: Jake Stauch says enterprise products require distinct development principles and product capabilities, and companies that opt not to build them leave the opportunity to competitors.

“And, and we've chosen to say like, "Hey, give me your your gnarly-est problems, and we're going to solve them and make AI, um, transformations possible in this, this crazy world and this world of risk and, and governance concerns."”

Claim: Jake Stauch says Serval has chosen to solve customers' most difficult problems and enable AI transformation despite risks and governance concerns.

“And, and we've chosen to say like, "Hey, give me your your gnarly-est problems, and we're going to solve them and make AI, um, transformations possible in this, this crazy world and this world of risk and, and governance concerns."”

Claim: Serval aims to solve customers' most difficult problems and enable AI transformations in environments with risk and governance concerns.

“Yeah. So when we deploy these software news, I think what's really important is we actually never build custom features. We don't build bespoke product capabilities, and luckily for our category, no one asks us to.”

Claim: Serval says it does not build custom features or bespoke product capabilities during deployments, and that customers in its category do not request them.

“Everything that people ask for is something that every other customer is going to want at some point. Maybe they impact the prioritization, but it's still something we've always wanted in the platform.”

Claim: Serval says customer requests generally concern capabilities other customers will eventually want; requests may influence prioritization but are already desired platform features.

“Now, because it is a code gen platform under the hood, what you build in Serval is very, very unique. So the password reset workflow for one company is gonna be very different, shockingly, than the password reset for another.”

Claim: Serval says its code-generation platform supports highly customer-specific workflows, such as password-reset workflows that differ greatly between companies.

“Um, same with onboarding and off-boarding and all these different automations internally. They look very different, but that's just baked into the core primitives of the platform. Serval almost becomes the IDE, right, for the IT department and other departments to build their automations and manage them at scale.”

Claim: Serval's platform primitives support customer-specific automations, and the company describes Serval as an IDE for departments to build and manage automations at scale.

“So what we do do is w- if we look and we see the same customer building the same automations over and over and over again and having to customize them and, and build them in the product in the same way, we start to consider, "Hey, should that be a first-class feature in the product?"”

Claim: Serval considers promoting repeatedly rebuilt and customized automations into first-class product features when the pattern occurs within a customer.

“So one example is we saw a lot of customers building workflows to detect knowledge base articles that were out, uh, conflicting or out of date.”

Claim: Serval observed that many customers created workflows to identify conflicting or outdated knowledge-base articles.

“And so we said, "Hey, we should just do that automatically." And we, we had customers that, you know, would, would set up automations to reclaim old licenses. So licenses for software you're paying for that no one is using, let's automatically cancel those subscriptions.”

Claim: Serval decided to make automatic detection of conflicting or outdated knowledge-base articles a platform capability.

“And so we said, "Hey, we should just do that automatically." And we, we had customers that, you know, would, would set up automations to reclaim old licenses. So licenses for software you're paying for that no one is using, let's automatically cancel those subscriptions.”

Claim: Serval customers used automations to identify paid software licenses that were unused and automatically cancel the associated subscriptions.

“Well, we just decided to build that into the platform. So it is a way for us to kind of test the waters of what should be really first-class features versus what should sh- stay as automations, uh, in Serval.”

Claim: Serval uses recurring customer automations, such as license reclamation, to assess which capabilities should become first-class platform features.

“Yeah, I think we're still watching this. I do see our CIO customers trying to pull back from that initial token maxing era, but I also think we're so early in the adoption curve that it's not going to be a, a clampdown in the aggregate.”

Claim: Serval is monitoring enterprise AI spending and sees its CIO customers starting to reduce spending after an initial token-maximization phase.

“Yeah, I think we're still watching this. I do see our CIO customers trying to pull back from that initial token maxing era, but I also think we're so early in the adoption curve that it's not going to be a, a clampdown in the aggregate.”

Claim: The speaker expects that early-stage AI adoption will prevent current spending pullbacks from becoming an aggregate clampdown.

“It's actually gonna continue to be a pretty dramatic rise because most companies are still figuring this out for the first time, and they're, they are all going to go through this cycle of rapid early adoption and then kind of a pullback of spend and trying to measure and calculate ROI.”

Claim: The speaker expects companies adopting AI to move from rapid early adoption into a pullback in spending and attempts to measure return on investment.

“It's actually gonna continue to be a pretty dramatic rise because most companies are still figuring this out for the first time, and they're, they are all going to go through this cycle of rapid early adoption and then kind of a pullback of spend and trying to measure and calculate ROI.”

Claim: The speaker expects AI spending to continue rising dramatically as most companies adopt and learn about AI for the first time.

“Um, but what I hear from our, our customers all the time is not really a concern about the, uh, long-term budget impacts.”

Claim: Serval says its customers generally do not express concern about the long-term budget impact of AI.

“They, they realize that there's gonna be a short-term pullback, but they're still planning on increasing spend. What they need, though, is more proof of ROI, and that, I think, is what's gonna transform in the industry is, is people looking for how can I prove that this spend is giving me something, and that is hard.”

Claim: Serval says its customers anticipate a short-term AI-spending pullback but still plan to increase spending, while seeking difficult-to-produce evidence of return on investment.

“I don't know that anyone's really figured out how to measure what you're getting from the tokens that you're buying, uh, but that, I think, is gonna be a huge opportunity for us to show the work that's being done and try to equate it to work that would've been done in other platforms or with other cost models or by labor, and so that's what c- our customer asking for is help me prove that this was worth it.”

Claim: The speaker sees a major opportunity for Serval in demonstrating work performed and comparing it with equivalent work done using other platforms, cost models, or human labor.

“I don't know that anyone's really figured out how to measure what you're getting from the tokens that you're buying, uh, but that, I think, is gonna be a huge opportunity for us to show the work that's being done and try to equate it to work that would've been done in other platforms or with other cost models or by labor, and so that's what c- our customer asking for is help me prove that this was worth it.”

Claim: The speaker is unsure whether anyone has developed a way to measure the value customers receive from purchased AI tokens.

“And, and that's what the assignment is. That's what we have to do. We, we try to price on workflows run because I think it's a better reflection of actual work being done versus just tokens, which is more like verbosity, um, and inefficiency.”

Claim: Serval tries to charge based on workflows run, which the speaker considers a better measure of work performed than token usage, which can reflect verbosity and inefficiency.

“Um, but, but I think we're all gonna figure out different models of proving that what we're doing makes sense.”

Claim: The speaker expects the industry to develop multiple models for demonstrating that AI work provides sufficient value.

“Yeah. I think all the companies I've talked to that tried to vibe code these tools internally, especially the, the main system record tools, not the, the kind of point solutions, but the main system record tools have discovered that it's quite expensive to build and maintain enterprise software.”

Claim: The speaker says companies that attempted to build enterprise tools internally with AI have found that building and maintaining systems of record is expensive, particularly compared with point solutions.

“Even if you can do it, it doesn't make economic sense in many cases to do it unless your needs are very bespoke and unique in some way. And so buying off the shelf for things that are not your core competency, I think has always been the build versus buy argument.”

Claim: The speaker says the build-versus-buy rationale continues to favor off-the-shelf software for functions outside a company's core competency.

“Even if you can do it, it doesn't make economic sense in many cases to do it unless your needs are very bespoke and unique in some way. And so buying off the shelf for things that are not your core competency, I think has always been the build versus buy argument.”

Claim: The speaker says internal development of enterprise software often is not economically sensible unless a company's needs are unusually bespoke and unique.

“I think that that is, is true now as it is, uh, as it, as it's ever been. Um- Because it c- still is hard to build enterprise software.”

Claim: The speaker believes the build-versus-buy argument remains as relevant as ever because enterprise software is still difficult to build.

“A lot, a lot of companies are discovering this or relearning this, that actually enterprise software is complex and you need people to maintain it.”

Claim: Many companies are recognizing again that enterprise software is complex and requires ongoing human maintenance.

“... uh, thing, things that, um, break down and people need to update it, so then you have to staff the team, and then you're paying for that labor and, and then no one really wants to work on the internal tools team in many cases.”

Claim: Maintaining point solutions can require updates and dedicated staff, creating labor costs, and many people do not want to work on internal-tools teams.

“So you have all these problems with doing that. Where I do see it is in kind of the point solutions where a lot of companies have hundreds and hundreds of SaaS applications. They probably need a couple dozen- ... total. And so I do think you're gonna have a shrinking of the, the sprawl of different point solutions, and they're gonna be eaten by the core systems of record and the core platforms.”

Claim: Companies that have hundreds of SaaS applications probably need only a couple dozen in total.

“... total. And so I do think you're gonna have a shrinking of the, the sprawl of different point solutions, and they're gonna be eaten by the core systems of record and the core platforms.”

Claim: The speaker expects the sprawl of point solutions to shrink as core systems of record and core platforms absorb them.

“Exactly. So I think there is gonna be consolidation. Um, and so this, there will be a SaaS apocalypse, but in that kind of point solution, um, smaller market cap type companies, I don't think that you're gonna truly get away from the courses records, the ERP, the HRIS, the ITSM, the CRM.”

Claim: The speaker does not expect companies to eliminate core systems of record or ERP, HRIS, ITSM, and CRM systems.

“Exactly. So I think there is gonna be consolidation. Um, and so this, there will be a SaaS apocalypse, but in that kind of point solution, um, smaller market cap type companies, I don't think that you're gonna truly get away from the courses records, the ERP, the HRIS, the ITSM, the CRM.”

Claim: The speaker expects consolidation in the software market being discussed.

“Exactly. So I think there is gonna be consolidation. Um, and so this, there will be a SaaS apocalypse, but in that kind of point solution, um, smaller market cap type companies, I don't think that you're gonna truly get away from the courses records, the ERP, the HRIS, the ITSM, the CRM.”

Claim: The speaker predicts a SaaS apocalypse affecting point-solution companies, especially those with smaller market capitalizations.

“I think those are going to stick around. They might transform, and what we're seeing with Serval is, is we are displacing the, those tools. Um, so I think there will be a new generation of incumbents, um, but I don't see those categories just disappearing.”

Claim: The speaker expects a new generation of incumbents to emerge and does not expect the core systems-of-record categories to disappear.

“I think those are going to stick around. They might transform, and what we're seeing with Serval is, is we are displacing the, those tools. Um, so I think there will be a new generation of incumbents, um, but I don't see those categories just disappearing.”

Claim: The speaker expects core systems-of-record categories to persist, although they might transform.

“I think those are going to stick around. They might transform, and what we're seeing with Serval is, is we are displacing the, those tools. Um, so I think there will be a new generation of incumbents, um, but I don't see those categories just disappearing.”

Claim: Serval is displacing some tools in the core systems-of-record categories under discussion.

“Yeah, absolutely. I think IT is really interesting because they're, they're closer to a software engineer than I think a lot of people give them credit for.”

Claim: IT professionals are more similar to software engineers than many people recognize.

“Uh, especially in, in large enterprises, the IT department is the engineering team. In a lot of cases, they don't have the, the software engineering team, so IT runs all these systems and, and IT often does write code and manage the systems.”

Claim: IT departments often operate enterprise systems, write code, and manage the systems they run.

“Uh, especially in, in large enterprises, the IT department is the engineering team. In a lot of cases, they don't have the, the software engineering team, so IT runs all these systems and, and IT often does write code and manage the systems.”

Claim: In many large enterprises, IT serves as the engineering team because there is no separate software engineering team.

“And so I think we treat them as adults. We give them the tools that we would give software engineers. Now, we abstract away some of that complexity, uh, we make it easier to use in some regards, and, and we don't just kinda hand them cloud code and tell them, "Good luck."”

Claim: Serval simplifies some aspects of its tools for IT users while retaining capabilities beyond simply providing raw cloud code.

“And so I think we treat them as adults. We give them the tools that we would give software engineers. Now, we abstract away some of that complexity, uh, we make it easier to use in some regards, and, and we don't just kinda hand them cloud code and tell them, "Good luck."”

Claim: Serval treats IT professionals as capable users and gives them tools comparable to those provided to software engineers.

“We actually, we build a platform that's dedicated to their use case, and we don't expect them to be able to necessarily write code or, or understand what's being written, um, but we still give them access to it under hood.”

Claim: Serval's platform is designed for IT use cases and lets users access code-generation capabilities without requiring them to write or understand the code.

“So you can always see the code.”

Claim: Serval users can always inspect the code generated or used by the platform.

“And you can always treat this through a s- full software development life cycle, but you'd also don't have to. And so a lot of the, the product principles is thinking with a few different personas in mind.”

Claim: Serval supports a full software development life cycle, but users are not required to follow one.

“You know, an IT engineer who actually can and does write code all the way through someone maybe in the marketing department or the, the people team that doesn't wanna see code ever, but still wants to have that flexibility of, of code gen under the hood because they've got a complex workflow that they wanna build that they don't wanna drag and drop their way to.”

Claim: Serval designs for both coding IT engineers and nontechnical users who want code-generated flexibility for complex workflows without interacting with code.

“So I think a lot of it just comes to having multiple personas in your head and also treating IT like an adult. I think a lot of the products that have been out there have actually ch- kind of been pedantic to IT of like, "Oh, you don't actually understand this stuff, so you need this very stripped down, more opinionated, simpler version."”

Claim: The speaker believes enterprise software should be designed for multiple user personas and should treat IT professionals as capable users.

“So I think a lot of it just comes to having multiple personas in your head and also treating IT like an adult. I think a lot of the products that have been out there have actually ch- kind of been pedantic to IT of like, "Oh, you don't actually understand this stuff, so you need this very stripped down, more opinionated, simpler version."”

Claim: The speaker says many existing products have underestimated IT users' understanding and responded with overly simplified, opinionated interfaces.

“Actually, IT appreciates the complexity and appreciates the power of these platforms. We wanna deliver that to them.”

Claim: The speaker says IT users appreciate powerful, complex platforms and that Serval aims to provide them with such platforms.

“Yep. So one of the things, I think every company discovered this as they were building kind of the first generation of these, these AI tools, is that you could build such amazing, complicated systems, but then you had to train people to understand all of your product primitives and how these systems fit together.”

Claim: The first generation of AI tools could be powerful and complex, but users needed training to understand their product primitives and how the systems worked together.

“And so the first generation of this is we deployed all these FDs to explain to customers, "Okay, here's, here's how to build in the platform. Actually, let me just build it for you." And then realized it would be really nice if customers could just talk to an agent that understood their requirements and then translated that into the product primitives.”

Claim: Serval's initial approach used FDs to teach customers how to build on the platform, but the FDs often built the systems for customers instead.

“And so the first generation of this is we deployed all these FDs to explain to customers, "Okay, here's, here's how to build in the platform. Actually, let me just build it for you." And then realized it would be really nice if customers could just talk to an agent that understood their requirements and then translated that into the product primitives.”

Claim: Serval identified a need for an agent that could interpret customer requirements and translate them into the platform's product primitives.

“And so we built this agent called Catalyst that does exactly that, is it an agent for managing the platform. So it's an agent for admins to build out all those underlying automations, but also configure the rules around approvals, uh, permission checks, um, build out different policies, update all these different systems, understand the, the entire deployment.”

Claim: Catalyst is Serval's platform-management agent for admins, supporting automation building, approval and permission configuration, policy creation, system updates, and deployment management.

“And so because Serval is all code gen under the hood, having the agent on top of that to help you manage the system just made a ton of sense and is so powerful because now I don't need you to train yourself on how to use Serval. You can just talk to an agent that actually understands how to use Serval and how to build on it, and, and that is your primary interface into the product.”

Claim: Catalyst lets customers use Serval through an agent that understands the platform and can build on it, rather than requiring customers to learn the platform themselves; the agent is the primary interface.

“And so because Serval is all code gen under the hood, having the agent on top of that to help you manage the system just made a ton of sense and is so powerful because now I don't need you to train yourself on how to use Serval.”

Claim: Serval's code-generation architecture makes an agent-based interface for managing the platform a natural and powerful fit, according to the speaker.

“Yeah. I always describe background agents to the companies like this is the ultimate moonshot in enterprise software, just these agents that go off and do work for you and magically discover all your problems and solve all your problems.”

Claim: Jake Stauch describes background agents that autonomously discover and solve enterprise problems as the ultimate moonshot in enterprise software.

“And it would be amazing if we could get that deployed, and it could work, and it could follow our permissions. And I think no one's really figured out how to do this well because on the one hand, you want these to be very autonomous, to go off and figure stuff out, and then solve problems without somebody babysitting them.”

Claim: Jake Stauch says deploying background agents that work within enterprise permissions would be valuable, but no one has yet figured out how to make them work well while balancing autonomy and safety.

“On the other hand, autonomy is very scary in the enterprise And so you do wanna see what exactly can they read.”

Claim: Jake Stauch says enterprise autonomy is concerning, and users want visibility into exactly what information agents can read.

“Because my best employees, what do you do with them? You say, "Here's general high-level mission. Go off, don't talk to me, and go do it. And you'll come back, and you'll have done amazing things." I want to be able to trust agents to do the same thing for me, where I can give them this high-level task, and they can come back and solve this problem and come back to me. I don't think we've figured out how to do that in a way that it feels safe- ... in the enterprise.”

Claim: Jake Stauch wants agents to handle high-level tasks autonomously and report back, but believes this has not yet been made to feel safe in enterprise settings.

“We think of it in, in the large enterprise, it's not perfect, but it's a, it's kind of an incremental escalation of capabilities. So you go off and you do these things, and it comes back and says, "Hey, I would like to be able to do all this stuff. This would help me do my job." And you approve them. And all the stuff are deterministic workflows. I wanna call this tool. I wanna run this script that does X, Y, and Z. Can I do that?”

Claim: Jake Stauch describes an incremental enterprise approach in which agents request additional capabilities and humans approve them step by step.

“This would help me do my job." And you approve them. And all the stuff are deterministic workflows. I wanna call this tool. I wanna run this script that does X, Y, and Z. Can I do that?”

Claim: Jake Stauch says agents can request permission for deterministic workflows, including calling tools or running scripts that perform specified actions.

“And you just kind of approve these with certain permissions and guardrails around them, and you say, "Okay, yeah, that's okay. That's okay, that's okay." And over time, the agent becomes pretty competent because it has a, a big enough toolkit that it can solve most problems.”

Claim: Jake Stauch says that gradually approving an agent's tools with permissions and guardrails can give it a large enough toolkit to solve most problems.

“Um, and it's not perfect because it's annoying to have to babysit that process, but I think it's what you have to do because otherwise it could theoretically just start writing code that could take action on all your systems, and that is an unbounded security risk that I don't think anyone's gonna be able to take anytime soon.”

Claim: Jake Stauch considers ongoing approval and supervision annoying but necessary, because unrestricted agent-written code could act across all systems and create an unbounded security risk that enterprises are unlikely to accept soon.

“I, I think it's, um, it's both, honestly. I think memory obviously has to be better and better because you wanna understand the preferences. You want the context to be insanely large because I don't wanna have a bunch of different chats with you.”

Claim: Jake Stauch believes agent capability depends on both models and their surrounding harness; memory must improve to capture user preferences, and context must be large enough to support a continuous conversation rather than many separate chats.

“I wanna have kind of one single chat that goes back all the way to our very first conversation, and you remember the key points of that that are relevant. Um, and then I think the permissions problem will always remain the biggest thing because even if I feel like you are a very, very valuable, smart agent, and I...”

Claim: Jake Stauch wants an agent conversation history that extends back to the first interaction and retains the relevant key points.

“I wanna have kind of one single chat that goes back all the way to our very first conversation, and you remember the key points of that that are relevant. Um, and then I think the permissions problem will always remain the biggest thing because even if I feel like you are a very, very valuable, smart agent, and I... You can do all these things for me, how much do I trust you to just run wild and not have to babysit your actions?”

Claim: Jake Stauch believes permissions will remain the biggest challenge for agents, since users must decide how much to trust even a valuable, intelligent agent to act without supervision.

“We are two hundred people today.”

Claim: Jake Stauch says Serval has 200 people at the time of the conversation.

“We were twenty in January, so it's been an incredible ramp year.”

Claim: Jake Stauch says Serval grew from 20 people in January to 200 people by the time of the conversation, describing it as an incredible ramp year.

“Yeah. The, the hiring velocity I think will, will increase even from where we're at today. Um, we believe that there is unlimited demand for work, and there's unlimited competition, and there's unlimited ambition.”

Claim: Serval believes demand for work, competition, and ambition are unlimited.

“Yeah. The, the hiring velocity I think will, will increase even from where we're at today. Um, we believe that there is unlimited demand for work, and there's unlimited competition, and there's unlimited ambition.”

Claim: Jake Stauch expects Serval’s hiring rate to increase beyond its current level.

“And so even though every single employee does so much more than what they could do a year ago, we have even bigger ambitions for what they could do.”

Claim: Serval employees are more productive than they were a year earlier, and Serval has greater ambitions for their capabilities.

“And so I, I think because every employee is so much more productive, we actually want more of them, not less of them.”

Claim: Serval wants to hire more employees rather than fewer because each employee is much more productive.

“I think what's so unique about Serval and the category we're in and the moment we're in is the TAM is unbounded. So it feels like the, the, the addressable market is really more about the size of our ambition than it is about fundamental limitations on how big this business can get.”

Claim: Jake Stauch views Serval’s total addressable market as unbounded, with its practical size determined more by the company’s ambition than by fundamental market limits.

“Because we are automating essentially all kinds of work for every single employee in the world, the TAM is as big as we want it to be, uh, and it's really how fast can we grow and what is our ambition and making decisions based on that versus, okay, here's, here's what the market looks like, here's what we think we can actually do.”

Claim: Serval aims to automate essentially all kinds of work for employees worldwide, and Jake Stauch describes its total addressable market as bounded by the company’s ambition rather than a fixed market size.

“And that's a different set of challenges. It's not, it's not, uh, strictly better because now there's so much open space, and there's so many more decisions we have to make about what we actually wanna focus on. Um, but it is really fun to not be constrained and, and hemmed in by what the market allows for.”

Claim: Jake Stauch says Serval’s open-ended market creates additional prioritization challenges, even though he enjoys not being constrained by market limits.

“We are embedded with customers. So even the, the engineers that are not necessarily deployed, they are in the customer Slack channels. We are talking about customers nonstop.”

Claim: Serval embeds engineers in customer communication by including even non-deployed engineers in customer Slack channels and ongoing customer discussions.

“They're hearing from customers every single day. Most engineers are on a at least one customer call a week. I think most of them are actually on at least a customer call a day. And so what this is, it's, it's like immersion in a, a country to learn a foreign language.”

Claim: Serval engineers hear from customers daily; most attend at least one customer call weekly, and Jake Stauch thinks most attend at least one customer call daily.

“You, you become so immersed in your buyer and your customer's mindset that you understand what they want, you understand how they're thinking about things, and you are adapting to their preference changes in real time.”

Claim: Jake Stauch says Serval’s close customer immersion helps employees understand buyers’ thinking and needs and respond to preference changes in real time.

“So as they're moving with the market and they're starting to talk about different products that they're trying out and things that they're excited about, our engineers are constantly aware of all of this and able to adapt very quickly.”

Claim: Serval engineers track customers’ changing product interests and can adapt quickly to them.

“So we've decentralized product development really to the engineering team. We don't have a product function because they're the ones talking to customers every day, and they're able to rapidly adapt to whatever customers are saying today and this week versus what they were saying a month ago.”

Claim: Serval has no separate product function and delegates product development to engineers, who use daily customer contact to adapt rapidly to changing customer feedback.

“Yeah. It's a big challenge. Our product fa- ships and changes faster a lot of times than anyone can keep up, including myself. Um, we just try to make sure that, one, the engineers that we hire are really product-oriented.”

Claim: Serval’s product often ships and changes faster than people, including Jake Stauch, can keep up with.

“Yeah. It's a big challenge. Our product fa- ships and changes faster a lot of times than anyone can keep up, including myself. Um, we just try to make sure that, one, the engineers that we hire are really product-oriented. That's one change we've made is they have to really understand our customers, and we wanna put them in front of our customers because they're gonna have to make so many decisions at the edge, and not every single product decision can be bubbled up to leadership every single time.”

Claim: Serval selects product-oriented engineers who understand customers and exposes them to customers so they can make many product decisions locally.

“That's one change we've made is they have to really understand our customers, and we wanna put them in front of our customers because they're gonna have to make so many decisions at the edge, and not every single product decision can be bubbled up to leadership every single time.”

Claim: At Serval, not every product decision can be referred to leadership for a decision.

“That is a fundamental change is you have to be willing to push a lot of the product decision-making down below, and you have to be willing to move fast and also undo and revert, uh, if you go in the wrong direction and, and kill your darlings.”

Claim: Jake Stauch argues that companies operating in Serval’s environment must decentralize product decisions, move quickly, and reverse decisions that prove misguided.

“You have to be willing to get... you know, change course on things that you thought were fundamental to the product and switch gears when your engineers bring you something that, "Hey, this is how the customer sees things now, and this is what we should do differently."”

Claim: Serval must be willing to abandon previously fundamental product assumptions when engineers report that customer perspectives have changed.

“The market's changed. The customer's changed. We're gonna do things entirely different." And, and I think that is what's enabled us to keep up is that ability and willingness to reinvent ourselves over and over again.”

Claim: Jake Stauch attributes Serval’s ability to keep up to its willingness to repeatedly reinvent itself as markets and customers change.