Redefining Energy Through Data and AI with Dak Liyanearachchi of NRG Energy

A traditional energy company is rewriting its future with data, AI and a customer-first mindset. We’re joined by Dak Liyanearachchi, Executive Vice President and Chief Technology Officer of NRG Energy, to explore how one of America’s largest energy companies is redefining its identity as a tech-forward, customer-centric organization. Dak shares how NRG is connecting the entire energy value chain — from generation to smart home consumption — through data and AI innovation. We also discuss the broader implications of grid efficiency, AI-driven forecasting and the cultural change required to power true digital transformation.

Show Notes [shortened for website]:

In this episode, Dak Liyanearachchi, EVP and CTO of NRG Energy, discusses how NRG is evolving into a tech-first energy company. He shares how they’re using data, AI and smart home integrations to enhance the customer experience and drive digital transformation across the entire energy value chain.

‍

Key Takeaways:

‍

(03:04) NRG spans energy generation to smart home energy use across deregulated markets.

(07:05) Transforming the customer experience starts with managing home energy at the last mile.

(09:48) Smart energy usage requires intelligent management of consumption and generation.

(10:49) Data enables personalized insights to help customers choose the right energy plan.

(14:24) A forecasting model simulates usage and market data to guide five-year energy planning.

(20:00) GenAI boosts productivity and improves customer engagement through automated support.

(25:02) The challenge isn’t adopting GenAI — it’s clearly defining what to do with it.

(28:57) Ongoing transformation is fueled by a tech-first mindset and a focus on delivering value.

‍

Resources Mentioned:

‍

NRG Energy website

Renew Home

‍

‍

Dak: [00:00:00] if you talk to my CEO he will say, we are not an energy company. we are more of a technology company than an energy company. And for listeners that think about energy as being kind of this stage industry, and for me coming into it, is that the rate of transformation and change is truly incredible. Even in their four and a half years

‍

I've been here. 

‍

Anthony: Welcome back to Data Masters. Today we're tackling a concept that's often thrown around as a buzzword, digital transformation, but we're gonna peel back the layers and explore what it means in [00:01:00] the raw real world reality of a sector you might least expect energy. We'll explore how a major player, NRG energy is fundamentally transforming itself, partly through a bold move, the acquisition of Vivin Smart Home.

‍

And this isn't just about an acquisition, adding a new service. It's about an established energy company embracing a digital native. Strategy, leveraging data and fundamentally changing how it interacts with its customers. It's a fascinating example of how a large infrastructure focused enterprise is actively innovating and disrupting its own business model.

‍

Leading much of this fascinating shift and bringing a truly unique perspective is our guest today, Dak Leonie Dak. Professional background is particularly interesting because [00:02:00] as you'll hear in a moment, it involves experience in highly customer-centric industries like hospitality and retail before he transitioned into the energy sector.

‍

And it raises an important question, what does it take to inject that customer first mindset, often seen as a hallmark of the hospitality and retail industries into an industry that's historically been driven by large scale, generation and delivery. We'll also delve into DAXs insights on the role of AI in this energy revolution.

‍

So let's get ready to deep dive into digital transformation and disruptive technology as we welcome Dak to DataMaster. Welcome.

‍

Dak: Thank you. Thank you for having me. 

‍

Anthony: So maybe we start before we dig into the details of digital transformation listeners may not be intimately familiar with NRG and the business and maybe just share a little bit about the history of the company [00:03:00] and on, and also your personal history and engaging the business.

‍

Dak: Yes, sure. So NRG is a Fortune 500 company. We have about 16,000 employees across our organization. NRG now is more than ever focused on technology. That ranges everything. From generation of electricity. So we have 24 plants across the US that generate about 13.4 gigawatts of. Generation capacity. So we generate an electron. We have many brands in deregulated markets where we will sell that electron. so think of that in terms of some of the northeast areas and in Texas where the market is deregulated, and that's both in terms of commercial and industrial brand for companies and also residential. then we. Manage an electron. So we bought a company called vn which is a smart home company.

So we, we see the whole ecosystem [00:04:00] from that generation of electron to selling the electron to then managing that electron in the home through, through smart devices. And just to give you context in terms of our, residential footprint, we're in about 8 million households across the us.

‍

Anthony: So, yeah, and, and very much to where I started. NRG is about creating, moving and distributing electrons and Most energy companies, or I think the layman's understanding of the energy industry is that it's a heavily regulated industry. It's largely about standing up power plants and thinking about, like literally how do you create energy in an electron.

And yet. Your history and background is very much not that. So maybe again, briefly just share a little bit about your background and why it may not been necessarily an obvious move for you to jump to NRG.

‍

Dak: Yeah, sure. So my background is. Really a lot of my time in the retail space in grocery retail [00:05:00] where customer data was critical to customer strategies, whether it be better ads, better communications pricing ranging in a store. so I spent a lot of my career there. Uh, Went to Hilton as their chief data and analytics officer to set up that function so we could start to understand from a, hospitality perspective, how to improve customer experience, guest experiences. then I had a conversation with NRG that was really on the journey of improving the customer experience and driving that through better understanding, better data strategies. And so hence I joined NRG just under five years ago.

‍

Anthony: So, so let's sort of put these two ideas together. And you had mentioned the acquisition of Vivin Smart Home. That to me feels like a. Pretty interesting strategic move. And if I could step back, the business [00:06:00] challenge NRG faces is how to transform the business how to become a digital first business.

And that's really a disruptive change in the context. Moving from the core strategy being about these physical assets to the digital experience. And you use an acquisition, in this case, vivin to. I think kickstart that digital transformation. Bring in fresh talent, fresh thinking and a bunch of probably, I would imagine, very useful technology with regard to that.

And this is a, common pattern we see when a business wants to disrupt itself, it uses m and a in an acquisition as a mechanism of spearheading. Is that what was going on under the. Under the covers here at NRG or maybe talk through the, both the ac the question of how to disrupt an established business, but also what role acquisitions play.

And then we can dig into a little bit of detail around, you know, how [00:07:00] Vivian helps in a very practical sense.

‍

Dak: if I look at N'S vision, we really wanna deliver the best energy and smart home products services that transform the customer experience and. to your point, we as an energy company have those big generation capabilities. We have that, ability to sell electricity as power and gas as well. But as we think about the home and the electrification of the home and it's becoming more and more and the demand is becoming more and more, it's a really good kind of synergistic way of thinking about the final mile, the last mile in terms of. Managing of elect electrons and certainly in terms of peak demand, how we can be much more strategic and much more responsive to our customer needs in terms of, how we think about load within the grid, load within the home.

What that does in terms of how we make that better experience in [00:08:00] terms of, Services, products, or even pricing for our customers. And that's based on that data. And if you think about it, that ecosystem is a lot, that value chain. We have data all the way across, all the way across from how we generate that electron to how it is consumed in the home.

And joining that up is. And it's not just the acquisition. We've created adjacent partnerships, strategic partnerships with other companies. So we can leverage not just the capabilities that we have, but capabilities like the thermostats like Nest, which we can talk about later. And be truly able to about the grid and think about the efficiency of the grid.

And, and one of the things that. You know, It may sound a bit strange from an energy company, but the cleanest megawatt is a megawatt we don't use. So the whole strategy there is how do we become, [00:09:00] intelligent around reducing some of that, that and, and be more efficient.

‍

Anthony: So if I put words in your mouth for a second, the Vivian acquisition is or was, I should say, about extending your data reach. Closer to the actual point of consumption, to the real experience the customer's having on a day-to-day basis with their home, their energy usage, the appliances they have and allowed you to create this end-to-end data experience mean clearly.

You had a lot of great data about, generation now. You had a lot of very rich data about consumption. Is that a fair way of framing it?

‍

Dak: Yeah, because ultimately we wanna be driving smart energy usage. Certainly as we think about the demand that is, we see in our markets today and we'll see more and more, and I'm sure the listeners and you have heard much many, many things about AI and [00:10:00] data centers and the demand that we'll see in the future.

So we have to be much more intelligent. As an energy company around how we manage that consumption and that and that generation.

‍

Anthony: if you don't mind for those of us who don't work in the energy industry, like in a way we're all good consumers. Like we, we know what it means to turn on our lights and run our washing machine, charge our cars. But maybe share a little bit about the practical side of this. You're collecting an enormous amount of really rich information and data, again, across the value chain.

What does this look like in practice? how would I experience this as a consumer? And how would you as a business experience it from a business perspective?

‍

Dak: as a consumer, we'll start with that. It's really around a few things. If I think about data and the use of data in personalization, understanding, letting a customer know their consumption, what their consumption's gonna be so they can better [00:11:00] plan. In terms of usage over the, the whole amount of the years, certainly in Texas when the HVACs kind of kick in and, energy consumption is, used a even we've things like using a customer data platform to inform customers in terms of are you actually on the right plan? Because it's so easy for a customer to pick a plan and because there are a lot of different plans and then not be on the right plan based on usage.

So using that data to, to actually understand and help that customer understand are they on the right plan? What's their usage like, what's their projected juice is like. so that's one part of it. And then as we think about, energy consumption and, you know, this is partly today and partly tomorrow. If you think about when the grid is being used a lot in those heavy days of hot weather, yes we can, we, with our partnership with Renew Home we can start to help try and manage the kind of [00:12:00] cycles in terms of. cooling in a home and do it much more. you know, It's not, Hey, it's hot, let's just turn it up three more degrees.

It's let's use data to be much more intelligent on how we know the weather is gonna come and how we make it. So you don't see that, but in the future we should think more about micro shifting. Your refrigerated uses a lot of power. it doesn't necessarily need to use all that power all the time.

So how can you micro shift and as you think about all the renewable technologies that are coming in, how can I really start to make your home more economical as I think about solar battery and grid. And so there are different times that it would make more sense to use different sources of power. And that's driven by understanding a lot of. The sensors and the data to, to create that ecosystem in the home that will be much more intelligent around smart energy.

‍

Anthony: let's dig into that a little bit because I think that's exactly the connection and relevance. Point here, which [00:13:00] is it's really about the data.

And I think again, historically we might have thought about an energy company, it's like, it's really about how do I generate electrons at a lower cost or move them around efficiently.

if again, your point is no, it's about really understanding all of the data. About all of the points of consumption generation movement sources, but share a little bit about how this impacts your data strategy and really this data perspective on it.

‍

Dak: Yeah. So as I think about the data strategy, when I joined the company, what was clear was we, to your point, we have a lot of data. We have so much data, it actually works in silos. Like a lot of them, you know, we're a large organization. We have a lot of data, but it sits in departments. as you think about pulling the thread through the value chain, what you start to see [00:14:00] is you see a lot more value through that data. So as I think for example, about a supply and demand equation, you know, whether it's retail, whether it's hospitality, whether it's energy, depends. It doesn't matter what sector, you always have the same question about supply, demand, and forecasting. Pulling all that data. For example, you know, we have a, forecasting model that, runs simulations on 15 minutes of usage data across residential and our CNI business. We use weather data from 20 years to start predicting out. We start looking at price curves of what prices look like for power, what prices look like for gas, because we will, burn gas to generate power. What does it look like in terms of our hedging strategy? What's it look like in terms of where we think the demand is gonna come? And what we build out is a five year plan all of those different things to say, okay. [00:15:00] our risk? Where's our ability to hedge more?

Where are we long? Where are we short? Should we generate more? Should we buy more off the grid? And so that's where you see the power of data and collectively around that value chain, bringing it together. Whereas before it was multiple different business units, potentially looking at their own sets of data and doing the, I think. Now we've moved to the probability of outcome conversation as opposed to, I think gut feel, this is what's gonna happen. [00:16:00] 

‍

Anthony: I love that as an example. First there's a, I think a little bit of a cult of. Big data, which is just that more data must be better. I think your insight here, which is really important for people is it's not just about more data, but it's about kind of insightful, the right data at the right time, at the right moment and that often, very much to your point.

Cuts across many different data silos. So this idea that as a business, if I think, you know, if I'm thinking about pricing data to use your example, then I'm just gonna go get, this narrow sliver of data around pricing versus cost data. You know, In your example or generation data, and your.

Core point here is that no, it's important to look not just at the volume of data act, how much can we gather, but really thinking horizontally across those [00:17:00] different silos. And then let's throw the customer dimension in as well. Because I, again, I don't wanna put words in your mouth, but I think what you're saying is that one of the important ways of viewing that data is through the lens of the customer, which again goes back to your, your. Background and history, but is that, fair or is that right,

‍

Dak: That is fair. That is fair. I, you've gotta look at it both ways. One, from a customer, what does the customer in perspective and then look at it from, okay, what does the business do to enable the right experience for the customer? 

‍

Anthony: so thinking from the customer back also helps inform or explain the Vian acquisition, but also really thinking about, again, not just in the context of, well, more data, but like what's the right data to inform the kinds of choices you make in the business in terms of where you generate, but also the choices that we as consumers make in the home in terms of, when we're charging our car, when we're running our [00:18:00] lights, or when we're what the set point is on our, thermostat.

‍

Dak: Yeah that's right. Antonin. And just think about it this way, as I connect the data dots, I have 2 million smart home households that we can understand their usage and their energy usage in terms of smart energy. And then we can scale that to to the supply desk. You know, What should the supply desk do, because. Traditionally, I look at big CNI organizations, the big retailers of this world, and we do demand response with them, but we're taking this to the masses millions of households and then doing that again in terms of what does that mean in terms of supply and what should we do in real time or the next day that you can't do without being able to connect those data dots and then be able to scale that to an extent that it makes a material impact on the business.

‍

Anthony: So you mentioned this in passing, but I, I want to. Sort of dig into it a little bit. You mentioned [00:19:00] predictive models that you've been building. You've discussed this in the context of the five-year plan but also the next day plan, like what you're gonna do next tomorrow. Uh, I think there's a, there's something really interesting here.

I think. In general, when we talk about ai, we tend to, or there's this I think especially recently, a lot of conversation around generative technologies and AI as a way of chatting. I suppose talking with your home or talking with your energy company I'm sure there's use cases there.

But you seem to be pulling on, I think, a more traditional view of ai, which is as a way of helping to forecast, to manage the business, and then using this rich end-to-end data experience you've created. But share with me your view on advanced analytics and AI as it relates to this.

‍

Dak: Yeah, and we, use a lot of traditional ai. I mean, we've been using AI for many years, but if [00:20:00] you, if we go into more of the gen ai, I. What that is doing is really driving more of that transformation. Because do use Gen AI within our CNI business, so large organizations that will deal with us and we have gen AI bots that are actually building out understanding what our customers need that they can either auto. a response from a human being and in the future it'll be a, relationship where it may not need a human being, but we're seeing significant gains in terms of productivity and even customer engagement scores are going up hey, you are calling me. I actually know what you are calling me about. I understand the context, and I can give you a much better tailored plan. And what we're seeing is that's not. reducing our manpower, it's about actually saying our manpower can do more strategic value added activity and engage in better customer [00:21:00] relationships. Let's use gen AI to do all of the work that we had a, whole load of people just pulling stuff together so we could have that conversation. And what we're seeing is we're seeing a productivity output because we're now talking and being able to service more customers than we have on the, on our CNI side because we've got bigger throughput. And I think that does is also starts our organization to think about transformation. I. we should be able to think about, not how does Gen AI help a business process today, but this technology should be able to transform how we do business in the future. with a lot of our organizations right now, we're actually doing that visioning of 2030. So we're doing the visioning of what does the organization of the 2030 look like? then work backwards in terms of how do [00:22:00] we use data, ai, gen, AI technology enable that. So we are thinking truly transformational. And I think, that is gonna put us in a good stead. And, the exciting thing is also a bit of a worrying thing because the organization has suddenly realized what we can do and the power we have by thinking in transformative ways. You see a lot of interest across just about every part of our business unit and departments saying, yeah, we need to think about our transformation and what that looks by 2030.

‍

Anthony: that's interesting. I just wanna highlight something you said there is a theme that I think has come up a lot in this podcast and I think is also has the benefit of, some truth, which is a lot of these generative technologies are about taking the kind of boring, repetitive, non-value added work that we often ask many people to do and getting that off their plate so they can actually spend energy and time focused on the places where their skills and capabilities are most [00:23:00] valuable.

It sounds like that's been very much your experience. I think I've said this before on the podcast, but when asked, where is a place where we can utilize this technology most effectively? Thinking about the places where the people in your organization are doing the most repetitive, boring work is the best place to start putting the technology in place.

It's like, what's the most boring thing you do on a day-to-day basis? Let's think about, gen ai there.

‍

Dak: Absolutely. And yes, there'll be different areas of the business that will have some of that productivity improvements in terms of, needing less to do more. But actually where we're seeing it now is actually increasing in terms of our revenue and sales because we have more opportunity and more bandwidth. To do more Of that relationship, more of that ability to price and where we have those incomings having the capacity to do more. so we're not looking at this right now as a, Hey, we're gonna shave X percentage of the bottom line. What we're saying is [00:24:00] we're actually seeing this as a top line growth opportunity and we're seeing that as customer experience improvement. 

‍

Anthony: I was gonna ask you about cultural hurdles that you've experienced as part of this transformation. It feels like a very negative framing. sounds like, you are actually seeing this as a great opportunity and to your commentary before, visioning what. looks like in 2030.

And it sounds like there's actually great engagement with the organization around that effort. So maybe my framing of cultural hurdles is actually exactly the wrong way to think about it. Quite the opposite. you're thinking of this as a way of reinve imagining the business but share with me, but maybe.

Some of the challenges and opportunities you've seen in getting a business to think very differently about its customers, the way it uses data and the way each of the parts of the business are thinking about the opportunity space. I. 

‍

Dak: And I think, one of the [00:25:00] challenges everybody has and I talk to other organizations is we've set an expectation that gen AI is the thing you've gotta do. and it's a journey, but when we start that conversation with with our leaders and even listening to other organizations, it's not whether it's, Gen ai, what are you gonna do with it? What is it we wanna do? And that's really hard to articulate sometimes

Anthony: that's why I always say that's the technology tail wagging the business dog.

‍

Dak: Exactly. But the business is asking for and I've actually been in meetings where I'm trying to get to the business, what's the business question? And the business is coming with a technical answer, which is ai.

That's a hard paradigm to shift, but it takes time just to understand that because people understand that it is transformative. But you've gotta understand what you're trying to do with this transformative tool rather than it's transformative and here's a tool, let's do it. So to your point, yes, absolutely. We don't want [00:26:00] the tail wagging the dog. and I think the other part of this is tradition, whether it's gen AI, data. Is that cultural change because does have an impact on the things that people do for the better. You're always, or human behavior will say, well, that's what I've always done. Now you're gonna change it. And I don't want, what do you mean you wanna change the way I do this? And I don't need to do this anymore. I need to do this. So you really have to get your leadership engaged. business unit to drive down that change. 'cause this is about, again good analogy. Tail wagging dog. This is not a technology or a data implementation. This is a business change. And change management is so critical and for us, we're actually, you know, we have change management as we do these things, but one of my colleagues is actually building up A-A-C-O-E around change management. that we have it coordinated across the business. [00:27:00] We're standing up the enterprise transformation office, so we have a consistent way of thinking about architecture change management. process improvement because these are all hard things to do, and what we don't want is it becoming five business units or five business units and functions doing their own things, and then we end up with a absolute mess across a relatively large organization.

‍

Anthony: no, I mean, I think that's very sound. And again, in, thinking about data that cuts across silos, what I hear you saying is there's also an opportunity to think about managing and driving change across silos, across business units and across the business. Is that fair way of framing it?

‍

Dak: Yeah, I mean, I, been pushing for a transformation office because what we don't wanna do two things. One, I would not wanna do a a load of data and technology to do the same thing, but quicker I think that's just a waste of. [00:28:00] As we think about the capabilities and think enterprise and raise it up, we really should think about what does that mean across the enterprise?

And back to where we started, Anthony, with connecting the dots. Those dots still need to be connected regardless of what we do in terms of business process and the power of that business process is the connection of those data dots.

‍

Anthony: I think that's a very sound insight and again, bringing this full circle building and changing an organization, disrupting a. Energy company to become this digital business connecting the data dot from the end consumer that who's turning on and off his or her lights back to the generation capabilities.

That's a powerful shift in change. And I think as you point out, well, setting you up to lead NRG into the 2030s.

‍

Dak: Yeah. And I think, as long as we keep demonstrating value and driving value and showing the value across the [00:29:00] organization we're gonna keep that enthusiasm and the thirst for more. if you talk to my CEO he will say, we are not an energy company. we are more of a technology company than an energy company. And for listeners that think about energy as being kind of this stage industry, and for me coming into it, is that the rate of transformation and change is truly incredible. Even in their four and a half years

I've been here. 

‍

Anthony: awesome. Well. Zach, thank you so much for joining us on DataMaster. I think there's a lot to take away from both how to transform a business and the important role data plays in that, and, how you're casting your eye forward in terms of the future. So thank you for making the time

‍

Dak: Thank you very much for your time. My pleasure. Thank you.

 [00:30:00] 

‍