Discover why traditional walled gardens fall short and how open standards are driving the next decade of enterprise ERP innovation.
Emma Coppin
Multimedia Content Lead, Thought Leadership and Customer Stories
Workday
As IT leaders race to keep up with the pace of technological change, choosing the right platform can feel like navigating a maze of tokens, tools, and short-term fixes. In this episode of The Future of Work podcast, Rogerio Almeida, GVP of International Partner Ecosystem at Workday, is joined by Tod Sacerdoti, VP of Product Management at Workday and former CEO and founder of Pipedream, to discuss why the traditional walled garden model is falling short in an agentic world.
Together, they explore the non-negotiables for today's CIOs—from trusting your system of record and embracing open standards like MCP and Agent-to-Agent to balancing rapid innovation with enterprise-grade security and compliance.
Host: In a world of constant change, how do you make platform decisions that won’t become obsolete tomorrow? For today’s leaders, choosing the right features isn’t enough. You need a foundation built to endure.
On this episode of The Future of Work, we’re joined by Workday leaders Rogerio Almeida, GVP of International Partner Ecosystem, and Tod Sacerdoti, VP of Product Management and CEO and founder of Pipedream. Together, they explore how an open architecture can help partners build a truly future-proof business—before the market moves without them. Let’s hear from Rogerio and Tod.
Rogerio Almeida: I am here in Vegas with a very special guest. He's twice founder of two fantastic companies. One's called Bright Roll that was sold to Yahoo. And of course, Pipdream, that work they had just recently acquired. On top of that, he's actually a published author. And because he also had a lot of free time in his hand,
Rogerio: He decided to become an angel investor. And because he didn't have anything else to do, he also is the father of two young ladies and a family man living in San Francisco. It is my friend Tod Sacerdotii. Lovely to see you Tod. How are you? and how's it going? I mean, you know, this is your first DevCon, right?
Tod Sacerdoti: Yeah, I mean first of all the buzz and the excitement in the room here is just it you can feel it just walking in. But I would also say, you know, I've spent most of my time in startups and just the production value that a conference like this has is just incredible. So it's exciting to be here.
Rogerio: It's great. Look, I'm a bit starstruck, right, with all of the stuff that you how did you package all in? Two companies that you founded and then, time of course to write your book and we'll talk a little bit about that and all of these other fantastic things and you're here at Workday now.
Tod: Yeah, I mean it's been a wild ride. Silicon Valley is very exciting place, but super feel fulfilled and fortunate to be able to found two companies and excited to be part of Workday today.
Rogerio: Now, you know, before we actually start with the conversation that we're supposed to have, there's a couple of things I want to clarify with you. I know that your father, back in nineteen seventy-three, he was already doing a PhD in AI. I mean, talk about foresight. Can you just tell us a little bit more about that?
Tod: You've clearly done your research, but yeah, I was born and raised in Palo Alto. My parents were both getting their PhDs at Stanford, and my father did get his PhD in artificial intelligence in 1973. I think at the time there was probably less than 30 people working on it in the world, but nobody cared, honestly, for 40 years and now everyone mentions it and
Rogerio: Influencer.
Tod: I mean, he was involved in startups in the eighties as like a CTO technical leader, so it was something I was around growing up and something I was excited about pursuing, but artificial intelligence in particular just wasn't
Tod: You know, it wasn't what it is today and so it just didn't have that level of yeah attraction and buzz that it does in this market.
Rogerio: So now you’ve got two young girls
Tod: Twelve and fifteen.
Rogerio: and fifteen. Are they gonna follow your footsteps and their grandfather's footsteps as well?
Tod: Yeah,
Tod: I mean I'm not super prescriptive about what they work on, but I have told them that and I say this to every young person, you know, entering college, coming out of college, the most important thing is understanding the impact of AI on everyone's day-to-day life, their productivity, how it's impacting businesses and being the domain expert, you know, coming into Are they using it? They are actively using it. They both have subscriptions and I would say they use it far more than they thought they would. But there are very clear restrictions in school what you can and can't do, so there's a lot of guardrails around how it can be used.
Rogerio: We talked and and you wrote a book, but you wrote a book using tools available to you. So what was the purpose of writing the book and can you tell a little bit of story about that? Yes.
Tod: The context is in February of this year I decided I needed to learn everything about cloud code, cloud co-work, and anthropic because it was having just such a significant impact on my team, our company, I think the tech industry broadly. So I set out, I gave myself 10 days to write a biography on Dario and the history of Anthropic. Yes. And I published a 275-page book. It's called The Scaling Curve. And it's about the founding and the execution of the business of Anthropic to February of 2026.
Rogerio: And that's available on Amazon.
Tod: It is available on Amazon, sells you know thousands of copies a month. It was really like a research project, more than you know, trying to be a published author, but I wanted to learn everything I could about the company and about Claude Code in particular and I did I learned it. Well,
Rogerio: What do you call the author?
Tod: the author, it's a pen name, it's called Claude St. John, but I actually asked Claude when I was writing the book to name the book, to come up with the author's name, to come up with the author bio, and to come up with the author's photo. So all of the only part of the entire book that was not created by AI is the cover.
Tod: 'cause I just didn't like the covers. But everything else is done by it.
Rogerio: Really cool. Now somebody promised me a signed autographed version of the book, right? So I hope that that somebody doesn't Right,
Tod: I will hand deliver it to you.
Rogerio: okay, great. Let's make sure I've got a long memory. Yeah, right. Hey Tod, let's just talk a little bit about Pipedream. How did you come up with the idea of Pipedream? Because it's such a cool piece of technology.
Tod: Sure, so my prior company, which was called Bright Roll, and it was acquired by Yahoo, and at the time we had like a second engineering team, we called them the product hackers, and they were responsible for building what we would now call like workflows or automations to make the business more efficient and effective. And when we were acquired, we had about 175 of these sort of standalone workflows running that had tremendous business value. And when we started our new company, we said, what did we build at our last company that was really valuable, we knew people would pay for, and we knew we could build as a team.
Tod: So we kept coming back to this product hacker team, and the idea was basically can we turn that team into software? And that was the original vision for the Pipedream product.
Rogerio: and Pipedream, the name is really cool. So where did that come from?
Tod: Yeah, well in I think it was 2019 we were pitching a company called Chime, which is now a public fintech company, and the CTO of the company was, you know, we were talking about building these automations and it was going to all be software and there'd be connectors to all these different apps. And he said, you know, that sounds great, but that's a pipe dream, it's not gonna happen. And I remember rushing back to the office going, you know, is the domain available and how can I purchase it? And we we bought it off a like Colombian domain squatter and the rest is history.
Rogerio: That's amazing. Congratulations. I mean, it's such a great stuff. And of course, augments the value that we do. I work there massively, right? We're at a conference, there's a lot of people interested in technology. Technology is changing really, really, really fast. Everybody's sort of backpedaling a little bit, saying, my goodness, you know, there's so much for me to learn.
Rogerio: IT leaders and CIOs in particular, they've got a plethora of a huge amount of stuff that they can choose from to use, but everybody's working around making the right decisions for platforms in a way that there is longevity and not just wasting money and tokens and people. What are the guardrails do you think that people need to start thinking when they're applying AI technologies in general and the choices that they make?
Tod: Yeah, I mean the first thing I'd say is I have a lot of empathy for being a CIO in today's market. I mean it's it's wild times and things are are moving super fast. But in terms of the non-negotiables from my perspective, I would stop focusing on features and tools, which seems to be like the topic de jour in every CIO conference or conversation that I see, and start first and foremost. For most companies, data and context is going to be the moat for that business. And so the first thing you have to do is think about like the architecture, what's the foundation? Do you have a system of record?
Tod: That has your data and context in it that you trust. Because everything is going to be built on top of that. So that's kind of like the most foundational table stakes. I'd say second is, does that platform support open standards? Because it's not an island, right? These systems have to talk with all the other pieces of enterprise software that a company has and its customers have. So that architecture, that foundation has to support open standards. And third is, you know, these.
Tod: Most of Workday's customers and people listening to this podcast, they're running large organizations. They have deep security and compliance requirements, they have very complex business processes. Those things need to be like embedded in these products from the beginning. You can't just bolt them on after the fact and hope they're going to work. So, you know, I tell the CIO, if if you have a system of record you trust, you have open standards supported, and you have your business processes, security and compliance embedded, you're in a really good position to face this new world of agentic AI and what's going to happen over the next couple.
Rogerio: So you're saying they're walled gardens right now, they're not great because then it's very difficult to be of course open systems and the connectivity between the various technologies that need to be chosen, right?
Tod: Yeah, I mean I I've been hesitant to criticize walled gardens historically because walled gardens have built some incredible businesses. I mean Workday, as an example, Workday for most of its history was a walled garden and creating a walled garden where your software worked absolutely perfectly for your customers, you know, built double digit billion dollar companies and huge successes. But I think everyone would agree that as we look forward, the walled garden model is not going to be the sustainable model in the future because the number of people or things interacting with your product are going to move from
Tod: individuals to agents and agents aren't going to work in these walled gardens in the same way. We don't know what tools our customers are going to use. Are they going to use Codex or Claude Code or Perplexy Computer or something that hasn't been created yet? As a company, and I would advise this to all companies is you have to break these walled gardens down and support working with the ecosystem that exists today.
Rogerio: you gave an example now of several development tools that can be utilized for you to develop code and come up with new agentic tools. What are the business reasons why companies and organizations should be thinking about those, you know, open system standards for tools? Why are they doing that now?
Tod: Well, I mean I think first and foremost, open standards just allow an organization and the CEOs, CIO's organization in particular to move fast. Right? So standardization allows a company to say, we want to move into this agentic future for, let's say, a specific vertical, like HR or finance or IT, but they don't have to like re-architect the way they work with their internal systems and their external systems. Standardization just lets them, okay, we use this vendor for HR and this other vendor for some other tooling.
Tod: And we want those systems to work together and we want to implement something in days or weeks, but not months or years. Yeah. So I think at the core, standardization allows CIOs to move fast and that's that's the goal today.
Rogerio: Then you know you talk a lot about standardization. There's a risk of organizations adopting non-standard apps and apps that live in individual islands. That would be, in your view, really bad for the business, of course, because of the lack of connectivity and have you. So how can organizations avoid that pitfall?
Tod: Well, I'd say again, I go back to my prior comment. The challenge of being a CIO today is you are asked to move fast, but then anything you do that has any mistake, you know, you're accountable to that result. And so I think standardization, and I'll point to two standards in particular that Workday has really adopted and many in the industry have adopted. MCP was a standard that was released by Anthropic, I think, about 15 months ago. Agent to Agent was a standard protocol released by Google about a year ago.
Tod: These standards enable companies like Workday, and a lot of the things that were announced here at DevCon are supporting those standards to immediately build agentic solutions for our customers and allow our customers to build them on their own that work with their existing stack, whether that's an internal tool, whether that's another vendor they're working with, or frankly something that comes out of Silicon Valley tomorrow. And standards allow companies to feel confident that if I allow an agent to interact with my system, my existing business processes, my security, my compliance,
Tod: That's all embedded in these systems. So the standards just let me interact with them in a compliant way and that helps people move fast.
Rogerio: You touched on something that we've been hearing a lot during Devcon and a lot of conversations that we have today is speed. And a lot of the tools being put out there is you know to make sure that developers can develop quicker and more profitably, and you know that perhaps you can develop more intelligently, right? So that is, you know, needs to be balanced against risk, compliance, data integrity. So how do you balance those two things? Because everybody's pushing for speed, but at the same time there's all these other things that they need to worry about as well, right? So what's the balance there?
Tod: Yeah, I mean I think I would normally have said before we were acquired, you know, there's a spectrum of security and compliance and companies have to figure out where they are in the spectrum. But I think one of the things I've learned at Workday is, you know, we're working with the largest companies in the world. There is no room for implementing solutions that are not secure or compliant. That's just like a non-starter.
Tod: So I think I go back to my prior comment, which is if you have a system of record that has your business processes, security and compliance embedded in them, you don't have to worry about these same types of problems where I'm giving an agent access and I need to slow down because I'm concerned that payroll is going to be released or an HR review is going to be released. The nice things about these incumbent systems is they have those security and compliance like built into the product. So when we connect an agent, we're already compliant from day one with our customers. Security and compliance systems that were architected.
Rogerio: Tod, always a pleasure, my friend. Thank you so much. I'm sure that we can speak for ages here about this. And for those of you listening to this podcast, please make sure that you subscribe through whatever tool that you use to get your podcast to listen to the future work. So thank you very much for work there.
Report
A gap between effort and impact is the real ROI problem. Discover the solution.