Latest / The Unpack Show with Anandh / Ep-17: Grant Ecker | Founder of Chief Architect Network
Transcript
- TheUnpackshow: Hey everyone, welcome to the Unpack Show with Anand, where we unpack less ideas and experiences and leadership lessons. I'm your host, Anand, here. today we are going explore a fantastic topic with my friend, you know, whom I've known for probably more than a couple of years now. the topic is going to be the evolving role of the chief architect as a modern change agent, right? So that's that's a that's a f fascinating topic for me. if you look at as AI transforms industries, architecture is no longer just about technology. it's about enabling innovation, it's about building trust, it's about driving transformation, enterprise transformation, and more importantly, aligning technology with the with with the business strategy, right? That's going to be real important one. So our guest has been at the forefront of these ev evolution, if you will, through his work with the chief architect forum and network. I'm part of the network as as well, by the way. If you if you are not part of it, check it out. helping architecture leaders navigate the opportunities and also challenges of the AI era. So without a further ado, I want Grant to introduce introduce himself and we can dive into the topic. Grant Ecker: Sounds great. So thanks for the awesome intro. yeah, Grand Ecker, I'm a global chief architect and lead integration at Ecolab. And previously similarly a chief architect from Danaher, Walgreens Boots Alliance, Medtronic, and helped to build the architecture practice at Lowe's Home Improvement. Did a lot of architecture stuff as well when I was at General Mills. the two of us are part of the chief architect network, collaboration. TheUnpackshow: I can actually purposefully. Grant Ecker: of about six hundred and eighty of us, peer architect executives, mostly around Fortune five hundred scale practices. It's a five one C three nonprofit volunteer group. Two of us really enjoy contributing to. And if that sounds like you join us in conversations like this one. Really excited to dig in. The topic TheUnpackshow: Yeah. Grant Ecker: is certainly timely. TheUnpackshow: Yeah. Grant, I want to dive into the evolution of the chief architect, right? So if you look at for decades, the chief architect's role was primarily viewed as, okay, how do we look at the organizational technical strategy, right? Someone responsible for to get defining technical standards, reviewing solution designs, ensuring systems are scalable Grant Ecker: Uh-huh. TheUnpackshow: and secure secure secured, right? These responsibilities are really important, right? The role has changed dramatically, right? Given AI, cloud native architecture, data platform, cybersecurity, a lot of evalu evolution that happened over the period of time. So now I see the business leaders now expect architects to shape the enterprise strategy, influence investment decisions, enable a innovation through AI and whatnot, right? So you've been part of this network and you've been chief architect yourself for a long period of time. how do you define this role in this modern era? maybe in the AI era, right? What are the responsibility do you believe as a chief architect is really important and how it has evolved over period of last ten years or twenty years, if you will? Grant Ecker: You know, it's funny. I think that this playbook has been here for a long time. And it sort of develops more relevance and timeliness to move closer to the strategy as companies go through major transformation and pivot. When they get into more of a steady state, you focus more on technical and delivery architecture. When they're transforming, your TheUnpackshow: Yeah. Grant Ecker: business and your information architecture become more important. And frankly, We're gonna we're transforming a ton right now and it's not slowing down anytime soon. In fact, one of my friends likes to say things are changing faster than they ever have, and they'll never move TheUnpackshow: Yeah. Grant Ecker: this slow again. I think that's a hundred percent right. So when you look at the classic TheUnpackshow: Very true. Grant Ecker: architecture foundations, like it starts with just your basic solution architecture to support your major projects and programs to make sure you're executing. In a way that's aligned with the enterprise design and that you're ultimately building things in the right way. so are you building things right? And generally, TheUnpackshow: Yeah. Grant Ecker: that's how a lot of architecture practices start even today. They start with suboptimal outcomes that might be identified across an organization and they go, well, geez, we don't want that anymore. What do we do? Well, let's bring an architecture team in to make sure that we fix that problem. So we get we get deployed into solving out. Things that are broken. And they well, that was effective. Maybe we should embed you into the process sooner and sooner until you're baked into a PMO TheUnpackshow: Yeah. Yeah. Grant Ecker: where when things come into intake, you're assessing them for architecture significance. What's the impact of this thing? Why does it matter? And then ultimately doing a tiration of how much engagement you need based on just how significant that is before something moves into execution. But that's just the beginning. That's just turning the creative technical architecture. You might start to pivot a little more left doing things like aligning your IT landscape. You know, do we have the right applications? What does our application rationalization strategy look like? What is our current state? You TheUnpackshow: Yeah, yeah, yeah. Grant Ecker: know, how do we look at our vendors and our partners and our standards and patterns that we want to deploy? How do we develop roadmaps and how do we develop reference architectures? Now that might all sound really proactive. Bad news, you're still. just a technical architecture shop. You're just looking at the technology, TheUnpackshow: Yeah. Grant Ecker: the how do we build things right. If the real fun comes in, what are the right things to do? A totally different question. That is where you're helping to clarify and articulate business strategy, align the key capabilities that are required. Might be really cool things like product and service innovation as a capability in your RD area. But it might be less TheUnpackshow: Yeah. Grant Ecker: exciting things like email campaign management. And key opinion leadership engagement. Why do those things matter? Well, if you work in a med device company and you actually have product and service innovation, you have an amazing new pacemaker or medical device. If you don't have the doctors in your industry talking about that device, it doesn't matter. It's it's it's actually TheUnpackshow: Yeah. Grant Ecker: never going to be heard about. So you have to take the holistic capabilities that you have and understand how those support your business strategy. And then assess your maturity. Like, are we really already great at RD, or is it something we have to build our capabilities on? Are we emailing people all over the place? And we need to be consolidating those to show up like a single company. Then your email campaign management becomes important. Determine that future state and TheUnpackshow: Right. Right. Grant Ecker: then create your roadmaps. Now you're really up front. There's still one more step. You got to build the information architecture, the things that are ultimately the data foundations that are required. What are the data gaps? to deliver on that roadmap. How do we determine our future state? How do we integrate with the roadmaps? And now you connect that to your technical architecture. And now you're aligning and overseeing the execution to the business strategy, the information outcomes, the overall technical landscape. And that's the pivot. And honestly, Anand, that's been the pivot from the beginning. Like architecture TheUnpackshow: Yes. Grant Ecker: shops that are really mature even 10 years ago, 20 years ago, did that same work. But just today it's so much more common to see an architecture practice gravitating graduating out of the basement and ensuring that we're just building things correctly to joining the business and understanding what are the right problems to solve. TheUnpackshow: Yeah, I think I see a lot of energy in you, Grant. Like, you know, I know it I know I'm gonna I'm gonna love this show. so I want to pivot one of the most valuable, right? I I myself was an architect. I still have the architecture mindset, right? one of the val most valuable assets for any leaders is creating opportunities for peers. And and you are doing that within your organization and even outside of your your organization, which is through the chief architect forum and network, right? I've witnessed it personally by myself, right? Where the architecture leader comes and you know, they are they're coming in and then they're discussing openly about challenges, sharing best practices, and then they're they're exchanging a lot of ideas that I've seen by myself, right? but communities like these help leaders stay ahead of the emerging trends, right? So with that in mind, I want to get to a point, right? What inspired you to create and grow this network, right? The forum that you created, which is the chief architect. forum and network. And what are some of the biggest themes that you're hearing from your you know, leaders within this group? Grant Ecker: Well, I created the Chief Architect Network because we needed to understand w where technology was moving and how we could be effective in using that that honestly massive change that we're seeing in the innovation speed to drive in to drive innovation and outcomes for our companies. And if you wait to read something about how what's gonna work, you're six to nine months behind. Because it had to be tried, it had to be tested, it had to be proven, and then someone decided to write an article about it and corporate communications approved it. So we're talking about a significant TheUnpackshow: Mm-hmm. Grant Ecker: lag. By then you're looking at one or two more waves of change in front of that. So that doesn't work. Similarly, a lot of the research organizations out there publish papers. And again, they're great when they come out, but they aren't really right in the here and now, which is why peer connects are so important. This is why TheUnpackshow: Yeah. Grant Ecker: a group like the Chief Architect Network is so critical because we can talk to our colleagues here and now about what's working, what's not. Take problems like AI FinOps, which is really hot right now. Understand what are the foundations you have and need to have, to have the the the we'll call them the utility meters outside of each house or each project inside of your organization, each each product TheUnpackshow: Yeah. Yeah. Grant Ecker: inside of your organization. So you don't have cross-billing confusion. TheUnpackshow: Mm-hmm. Grant Ecker: And then managing. The utility bill within each of those houses is is just as important. But if you just have a neighborhood meter, it's impossible to drive accountability and plan your your your operational costs on AI and cloud consumption. So coming together as peers, we're we're now meeting once a once a month on Fridays talking about this is incredibly helpful to find our way to what what's working before somebody can publish the hero case study. three to six months from now, that might have some compelling answers that really we needed to have this summer. So that's TheUnpackshow: Right. Grant Ecker: really what the impetus behind this network is. At the same time, we can advance our profession by having architect leaders on stages and sharing thought leadership and build our our connect connection of and access to talent. We are at different altitudes. We generally start at the director level and move up to SVP and beyond. And that allows us to really have access to the other strategic thinkers in the network both to find our next job or to hire the next talent that we might need in our organization and also help each other along the way. So TheUnpackshow: I think we could have a question. Co. Hey Grant, that was good. so I I want to pivot into given the AI transformation, every organization, right, wants to become an AI enabled, but many initiatives fail to move beyond pilots. I think we have seen that in the last year and even some of the cases that we are seeing even in this year. but the challenge is always about fragmented data, weak governance, unclear ownership, unrealistic expectations. There's a lot that you can keep on adding to it. But when I look at from a chief architect or an architect perspective, you're not here to just do the solution. You are here to influence the leaders. You are here to bring the other architects and the engineers along the journey, right? So given you cannot just say it's a fragmented data, we cannot do it. It's a weak governance, we cannot do it, right? So from your experience, what are some of the most architectural mistakes that organization? make during AI transformation specifically, right? we have seen digital transformation, we have re seen a lot of transformation, but specifically on AI transformation, and what would be your practical advice? Grant Ecker: Well, I think it's where you're trying to solve the problem on the river of the the solution. So if you ultimately have tried to deploy a vertical AI solution for a specific use case, and you then expect it to scale, that's that TheUnpackshow: Yeah. Grant Ecker: doesn't work. So I think a lot of people come back with the data's not right or the governance isn't right. But actually I would say you've solved a problem down in Louisiana. when you needed to be solving a problem in Minnesota at the St. Croix. TheUnpackshow: Mm. Right. Grant Ecker: so it's it's that's the issue is you didn't design a horizontal look at what you needed to be true to create an agentic ecosystem. something that could be enabled by decisions that are made on the edge or ultimately have the decision intelligence. To be able to represent our logic on AI's now ambient reasoning edge. So TheUnpackshow: Right. Grant Ecker: if you haven't done that work and you effectively are sort of just doing either classic machine learning or you're using a rules-based approach, of course it doesn't scale because you were really looking at a problem and curating by human effort how you basically do a slightly better pivot on robotic process automation. TheUnpackshow: Yeah. Grant Ecker: with AI, right? And you really weren't taking a 3D chess approach to the problem, which is what's necessary. You now need to think about how can I teach a machine to think, act and learn. All three things. TheUnpackshow: Yeah. True. Grant Ecker: Think is one step, act is another step, learn is another, and then the how am I going to re-architect the TheUnpackshow: Yeah. Grant Ecker: the landscape of permissions and trust. Which means I need to know what the agent's allowed to do, what it's not allowed to do. think about the bowling metaphor. This is one I like to use a lot. If you're bowling and you're not very good at it, or maybe a little chaotic, you might end up with the ball in the neighbor's lane. You might end up with it all over the place. So how do you get the bumpers that keep the ball in the lane? And then how do you narrow those bumpers over time so that you're throwing strikes every time? and TheUnpackshow: Right. Grant Ecker: how do you start to adjust permissions and trust? So that the guard that's blocking the pins is raised and the ball, which is headed in the right directory, has access to actually take actions and knock over the pins. And when you don't get them all, how are we learning from that to adjust how the bumpers are set up or how the ball's trajectory is thrown? So that could be the permissions and trust, the the bumpers and the and the front, or it could be how the ball itself, the agent, is acting. And ultimately that is the iterative experience of moving to a scaled AI approach, is you're taking the decision intelligence into the agent and the governance intelligence into the guardrail, deciding what the agent can and can't do, because the TheUnpackshow: Right. Grant Ecker: agent might decide it needs to delete the entire code base, like we saw at Amazon, and take the system down for seven hours while we do a DR, right? And the reality is it should have never been allowed to take that action. There should be governance over the agent's actions that say, Are you allowed to take that act before it can happen? Because you don't trust the agent, you trust its controls, you trust what it's allowed to do. So ultimately, that is what I think is what's preventing today's scaling. It's it's a root issue TheUnpackshow: Yeah. Grant Ecker: of addressing at the bottom of the Mississippi the trash that's coming to that to that delta. As opposed to going up to the Saint Croix and saying, What garbage are we adding to the river? TheUnpackshow: Yeah. Yeah, I think you brought up an inter interesting point, right? about AI agents. I think we are entering into a we are we are already or we are entering a future where AI ag agents can well collaborate with humans, autonomous agents will make necessary, you know, complex decisions and enterprise technology landscapes will continue to evolve for sure. I I'm excited for the architects, right? You know, I I I I tell my team that, you know, hey, we are no longer going to be engineers, we are going to be designers, right? Because we need to design the agents and the agents will perform what it needs to perform from a goal and and even from a coding standpoint, right? So it's exciting for the architecture and the architects. But looking ahead, you know, probably five years from now, Grant, or maybe two years from now, right? Things are rapidly changing. How do you see the role of chief architect is evolving? and and if you want to succeed in the AI first world. that we are heading into, that we are already into. what is what do what what are your thoughts on that? Grant Ecker: That's a great question. So, again, I think if we zoom out to the fundamentals of architecture, it hasn't changed. Business architecture and strategy and execution has been a part of our playbook for a long time. How we execute information architecture and and and technical architecture needs to pivot to cover this paradigm shift. Just like the printing press and and moving to connectivity with the internet and cloud, the the ambient availability of reasoning is a is a change that requires us to re-architect how we actually design solutions to be able to think on our behalf. So that's the biggest change, I think, is that it's not just a technology deployment, it's a logic deployment. And that's a big pivot. Other than that, TheUnpackshow: Yes. Yes. Grant Ecker: though, the and I think the governance and all the things we talked about applies well. But other than that, I don't want to pretend that. There are no rules, and the rest of what we've learned all this time doesn't apply. It still very much applies to understand your TheUnpackshow: Yeah. Right. Grant Ecker: business problem, the capabilities required, the gaps that we have, the roadmaps to get there, along with the targets that we have in mind, and the patterns to kind of stay in a standardized environment. so that we don't end up with a bit of a a cobweb of of how things are done. Those fundamentals remain true. But what's interesting is like every other profession, we need to focus a little less on that detail of how and use AI to help us define that how and oversee it, where we can focus more on the what are the right things to do and how do we decide what is right, which is really logic codification that is the is the pivot. And we need to encourage all of the other architects in their individual domains and technical leaders in their own domains to do the same thing. Become the overseers of their area of expertise and to understand how their area connects to the areas next to it. Become more T-shaped. including the business, including TheUnpackshow: Right. Yeah. Grant Ecker: how the business connects to us. A friend of mine calls that pie shaped, where you add a you add another part that comes down from that top of broad understanding TheUnpackshow: Or shape, yeah. Grant Ecker: to communicate in business terms. I think that's a a really nice metaphor. I think that's the pivot, honestly. TheUnpackshow: Yeah, yeah. I think like I said, Grant in the beginning of the call, I know you are you have a lot of energy towards this topic. you know, I was excited for this topic and I was like thinking about what topic I should talk talk the talk talk talk you know, talk to Grant Ecker, right? You know, then this came up and then I I wanted to use this time to discuss this topic. I learned a lot of stuff, right? You know, like you said, the fundamentals are going to remain, the core patterns are going to remain, the core principles are going to remain. Looking forward for the the the next can the Chief Architect Network meetings grant. and thank you for taking the time to be part of the show and I wish you good luck. Grant Ecker: Thanks, my friend. Great to be here with you. I loved the conversation. I think you're thinking about where we're moving and maybe be lucky enough to get to see you down in Chicago September ninth through eleventh, when we all get together. great show and thanks a lot for TheUnpackshow: Well well, definitely, yeah. Grant Ecker: having me on. TheUnpackshow: Yeah, thank you so much.