Premji Invest | CEO Series | DOSS: Turning operational complexity into competitive advantage

Summary

What if ERP was built for how businesses actually operate, not just how they account?

In this episode of CEO Series, we sit down with Wiley, Co-Founder and CEO of DOSS, to unpack why legacy ERP systems have failed operations teams and what a new category, “operations cloud,” looks like in an AI-native world.

For decades, enterprise software has optimized for tracking dollars. But modern businesses run on something far more complex: the flow of goods, data, and decisions across increasingly fragmented supply chains. DOSS is rethinking this from first principles, building a system designed for operators, not just finance teams, and leveraging AI to fundamentally change how software is implemented, maintained, and scaled.

In this conversation, we cover:

Why ERP has historically underserved operations leaders
The rise of “operations cloud” and what it unlocks
How supply chain complexity has shifted for modern businesses
What it means to build an AI-native system of record
Why the future of software is agentic, self-driving, and outcome-driven

Wiley also shares a candid perspective on competing with incumbents like SAP and NetSuite, the evolving role of AI in enterprise software, and how DOSS is positioning itself for the next generation of computing.

DOSS recently raised a $55M Series B, co-led by Premji Invest, to accelerate this vision.

About DOSS:

Wiley is the Co-Founder and CEO of DOSS, where he is building an AI-native operations platform for companies that make, move, and sell physical goods.

About Premji Invest

Premji Invest is a global investment firm backing companies that are shaping the long-term future of technology, enterprise, and society.

‍Vedant: [00:00:00] Hey, everyone. Thank you so much for joining us. Uh, this is our next episode in our CEO series featuring Das, and I've got Wylie with us, the CEO of the company

Vedant: I'm curious, you know, you are in this category broadly speaking called AI ERP, right? But you're not calling it an ERP, you're calling it an operations cloud. So I'm curious to understand a little more about your vision about this new category-

Wiley: Yeah ...

Vedant: and the end user that you're trying to serve.

Wiley: If we look at the acronym ERP, it's kind of a non-acronym.

Wiley: It doesn't really mean anything. Enterprise resource planning. I don't really know what that means. And if we look at the history of each one of these, uh, especially from like a generational standpoint, we had these mainframes. IBM built these back in the late 1960s, early 1970s, and then you put applications for finance and accounting on them.

Wiley: '70s and '80s, we get [00:01:00] PCs, end of the '70s. Um, we start inserting disks and actually, you know, installing software on client server. That was the second generation. And then towards the end of the '90s, we found out that we could put these applications on the internet, and then eventually, you know, we moved from PeopleSoft to Workday, Siebel Systems to Salesforce, and we came up with the idea of the cloud.

Wiley: ERPs are amazing at tracking the flow of dollars. They are not amazing at tracking the flow of goods. Our ability to work with operations leaders and help them model the complexity of their business, which is growing a lot in the last couple of decades, and we were talking about this earlier, it's something that we've been able to give them a platform that they can grow with- That is independent of the rigidity and accountability that you want out of a standard finance and accounting system.

Wiley: So Operations Cloud takes procurement, inventory management, order management, freight, understanding the master data management, things that traditional ERP has not been amazing at, moving that out and connecting it into third-party finance and accounting systems in a way that is really thoughtful and [00:02:00] intentional.

Vedant: Is the operations getting more complex? Fast-forward 20 years ago, look at it today, the operations leader was m- mostly the second-class citizen, but they were making do with, you know, duct tape k- kind of measures with Excel spreadsheets and the like. So why does that not work anymore, and why does, why does an operations team say, "Hey, I need the solutions now"?

Wiley: It's never been easier to build a business, but at the same time, that means the competitive pressure is much higher. We meet super young companies that are doing even just, you know, a single-digit million dollars in revenue who have operations that are as complex as a company that's making $100 million in revenue, because they have to go out and compete in every single surface for customer attention, for customer love.

Wiley: Amazon has, you know, set a lot of expectations about same-day shipping and things like this. Like, it has never been harder to compete in today's landscape, therefore, the complexity of the business has to match that, and so they need tools to be able to manage that complexity.

Vedant: You know, then I'll be curious maybe to dive then into DaaS, right?

Vedant: So how do [00:03:00] you guys solve it? I know there's a, there's this notion of a flexible data model that you've built out, but give us some color on what is built here, why is it so complex? Uh, why is it so hard to do?

Wiley: It's really hard to be able to do flexible things at runtime inside of a database, and then also make it behave like a database and be transactional.

Wiley: So what we actually ended up building was functionally a, you know, hybrid transactional analytical processing database, HTAP. The, the moment our applications touch this system, it feels like they're touching the raw performance of a database, but with the composability of something that feels a lot more like Excel.

Vedant: Are you trying to rip out the ERP? Are you doing something different? How do you ensure the operations leader can use the software when it still works well with the GL? What does that vision look like for you?

Wiley: Yeah. We're, we're definitely not trying to rip out the ERP. Businesses have grown in their complexity, and as they stretched, they stretched their ERP into places that it really doesn't make sense, and that's left [00:04:00] space for a solution.

Wiley: And that's, that's really what category creation is, is like the market expands and it, it-- and holes emerge where you're like, "Hey, I didn't realize there was a hole there before. We should probably fill that with something. What do we put there?" And our whole p- that's our whole point with Operations Cloud, and it, it's something that has probably been needed for a while but wasn't technically feasible before.

Vedant: So help us understand. You know, I'm sure there's a lot of customers watching this. Walk us through what, what do you guys do, say, in the first week of entering a customer?

Wiley: I think the thing that- is most interesting to me around talking to our customers is they mostly are interested in capabilities.

Wiley: They'll come to us and say, "We really want demand planning." And, you know, we, we can be a bit, uh, fun with them and be like, "Okay, what do you actually mean by that?" And then they explain it. You're like, "Okay, so you wanna understand how you can go out and automatically produce enough inventory to fulfill the orders that you think you will have at some point in the future.

Wiley: But that forecast for demand isn't known. The [00:05:00] ability to understand current levels of inventory isn't known. Your ability to reconcile the SKUs that you purchase, the SKUs that you store, and the pla- SKUs that you sell also isn't known. So you want demand planning, but we're missing, like, five things in between that."

Wiley: And I think, um, if you remember, you know, Maslow's hierarchy of needs, right? Like, sometimes customers are, like, coming to us and asking for enlightenment. I'm like, "When was the last time you ate? You don't have food, water, shelter. Okay, cool. Let's start there," right? Our sales leader would, you know, would kick me for not saying this.

Wiley: The only things our customers care about, how do I improve the profit- profitability of my business? How do I grow my revenue? How do I meet my customers where they are and make them happier? And how do I mitigate future risks? We don't have to, you know, uh, read the tea leaves to understand these businesses.

Wiley: Now, what we do need to understand is we go in and go, "Okay, how do you do that as a company? What is unique that you do? Oh, you're a coffee business. You care a lot about procuring green coffee from your place of origin in Ethiopia [00:06:00] or in Venezuela or Colombia. Oh, and you actually buy it directly from the growers, and you wanna track the lots from the growers.

Wiley: Okay, that's unique. We have to figure out how to massage that in and adapt that into our current procurement process that we run as a default."

Vedant: I'm curious, you know, there's one question that has been plaguing software investors over the last couple of months, right- Mm-hmm ... which is AI will eat software.

Vedant: And the concerns, if I may, as an investor, are broadly threefold when it comes to AI with application software. The first concern is that with vibe coding, I can vibe code many applications already.

Wiley: Mm-hmm.

Vedant: I don't have to get it perfect. I just have to get it 80% good enough. And if it's 80% good enough, maybe that applies some amount of pricing pressure on the incumbents.

Wiley: Right.

Vedant: The second issue is what if these AI labs move up the application stack? Right. And if I'm a user and I'm using the AI application to use the app- the SaaS application on the back end, then guess what? I need fewer seats on the back end-

Wiley: Which- ...

Vedant: directly impacts our revenue model.

Wiley: Right.

Vedant: And the third is that, hey, if I have now [00:07:00] become the system of engagement, I being the frontier lab-

Wiley: Yep

Vedant: um, then does the application at the back end just become a dumb kind of data pipe? How does DaaS mitigate against some of this? Are you guys working with this tailwind? Are you working at as, as a headwind? How do you think about the business over the next three to five years?

Wiley: In the analogy of headwinds and tailwinds, I think there's an omission of the fact that you can turn the plane And like, you like, you can choose where you go.

Wiley: And I think that's a, that's actually the story that's happening right now in the market is like, do you want AI to be a headwind or a tailwind for you as a company? It comes down to these periods again, of these generations of technology. Gen one mainframe, gen two client server, gen three cloud. Our view is that we are in a fourth generation of transformative change, and so we're gonna orient the company to benefit from AI eating software.

Wiley: For us as a company, that means that you make a couple of different decisions around product and, uh, engineering, but then also around how you build your revenue model and how you scale the [00:08:00] business. So like, we're not gonna sell to users. We know that the users of the future are machines. We need to sell infrastructure that they can consume.

Vedant: Let's fast-forward in a time where the human is now becoming the reviewer, not the doer.

Wiley: Right.

Vedant: What could this new paradigm look like and what is, how do you build a product for it?

Wiley: I think managing AI And agents in general, it's gonna be a lot like managing people and managing organizations. And ask any manager, they're gonna be like, "My best employees are the people who bring me problems, and they tell me solutions, and they manage up to me."

Wiley: Systems are gonna drive themselves. I think that is the hallmark quality of what is going to be the Gen4 architecture, is that you have self-driving software. And the same way that you... You know, Uber was an amazing innovation. You can click a button and a human being comes and brings you a car. Waymo is a fundamentally different thing.

Wiley: You click a button and the car brings itself to you. And I think that the further extension of it is that the software will deliver to you packaged up [00:09:00] answers to problems you didn't even know you had, and solutions that it partially delivers or fully delivers depending on, uh, what it assesses the risk profile of those things to be.

Vedant: You also have a term that you use extensively, which I like a lot. It- you call it the ERP racket.

Wiley: Hmm. Yeah, yeah.

Vedant: Um, and you basically say that a lot of the vendors that exist in the market today are actually incentivized to make their product as complex as possible to use. Yes. And so that's a pretty big claim that you've made.

Vedant: Uh, can you kind of expand on it, and how is DOS different from that regard?

Wiley: I say it a little bit in jest, but like, you know, in, in every joke there is, like, some shred of truth, right? If I were to think about our own infrastructure and our own companies, it'd be like taking the most important, you know, system design of how your business functions and saying, "I'm gonna give that to somebody else.

Wiley: I'm gonna lock it in a box that I don't understand, and I'm gonna ask this person to basically, uh, you know, uh, charge me money every single time I wanna open it and understand it." Um, which is kind of a fundamentally insane claim, you know, on, on, on its face. But the benefit of that is, is that they take on the [00:10:00] complexity.

Wiley: They do the hard work. And so it has its pros and cons, and our view of this is that the role and relationship of the systems integrators to their customers and their software vendors is gonna change dramatically. They're gonna get pushed up the stack in a, actually I think a materially good way for their businesses.

Wiley: I actually see this as a benefit for everyone in the industry, which is that a lot of the work that is mundane, difficult, taxing, very hard for the organization to operate, that's all gonna be done by agents. It's gonna all be done by machines. It's gonna be manufactured And the work that needs to remain is how do we manage our relationships with our customers?

Wiley: How do we deeply understand their business? How do we work with people to understand the things they want, the things they need, and things that they wanna go do in the future and, like, the business objectives?

Vedant: You know, you're going up against some pretty big wits. Yeah. You're going against SAP, NetSuite, QuickBooks.

Vedant: I mean, these are large companies, billions of dollars of revenue. Yeah. Very entrenched. How does a small company, you know, just a few years old, take them on? And, you know, why should an engineer who [00:11:00] has so many options, right? Whether they wanna join some of these companies or, say, the AI labs or something else, uh, why would they join you versus something else?

Wiley: The thing that we're typically competing against is actually inaction and the business' inability to execute. Very rarely are we directly competing for I'm evaluating vendor A and I'm evaluating DOS. The reason that I think this is exciting from a technology standpoint and why people, you know, in the battle of the David and Goliath here is like, this has happened before.

Wiley: Every time there is enabling technology, that always happens at a low level, foundational level. There's this lag where the enabling technology gets brought to market, and then people are trying to figure out, "What do we do with this thing? What do we do with the operating system? What do we do with the internet?"

Wiley: And it takes 10, 15, 20 years to actually commercialize it in its full true manifestation, which ultimately a lot of times is at the application layer. My view on it is that it's a bit of a arms [00:12:00] race right now. How fast can we build the agentic applications company and the agentic system of record, uh, versus how fast can the incumbents rewrite their system to be agentic?

Wiley: That's really the next question is, how fast can they get to that, or how fast can we get to it? I like to bet on small startups being nimble, so.

Vedant: So do we. Yeah. When you were in the market, how did you decide who the right partners were?

Wiley: You have to find- A group of people who believes in the world that you're building toward, and will irrationally work through those problems with you.

Wiley: You were to go to a public markets investor, you know, someone who spends all of their time looking at, like, the precise ways that the strategy of the company could play out. By every shape and form factor, it will, you know, uh, from a probability standpoint, it's gonna fail. But when we spend time with, you know, teams like you guys, it's very clear that you're working to make the low probability thing happen, and partners who believe in you, they'll go do that.

Wiley: People who are there because, [00:13:00] you know, oh, you might be the cool company that's in market right now, and this cool thing's happening, those are the people who go away when things aren't good. And you... That's the last thing that you want. You want people who are there, you know, when you send bad updates, and you're like, "Hey, things aren't going well.

Wiley: I need help." The people who are the first people to jump in are, that's who you want, um, and that's what you have to look for.

Vedant: Well, Wally, on, on behalf of everyone at Prime G Invest, I, I can tell you that we are stoked to be in business with you. Likewise. Um, and we can't wait to see what, uh, what Das builds in the future.

Vedant: Awesome. Thanks for doing this. Of course.

‍

CEO Series | DOSS: Turning operational complexity into competitive advantage

Recommended

Formal verification just left the lab. Why we invested in Pramaana Labs.

Formal verification just left the lab. Why we invested in Pramaana Labs.

CEO Series | Enveda: Building a 21st century generational pharma company.

CEO Series | Enveda: Building a 21st century generational pharma company.

CEO Series | DOSS: Turning operational complexity into competitive advantage

CEO Series | DOSS: Turning operational complexity into competitive advantage

Why I take founders on a 3-mile hike before writing a check

Why I take founders on a 3-mile hike before writing a check