Podcast / CXO of the Future

Maria Latushkin, Group Vice President of Technology and Engineering, Albertsons

Today we welcome Maria Latushkin, a technology and product executive who has spent over 20 years leading Product, Engineering, and Operations teams transforming technology for global brands including Walmart, Peet’s Coffee, and most recently Albertsons.

As Group Vice President of Technology and Engineering at Albertsons, Maria is responsible for all aspects of technology for retail, payments, and supply Chain, thinking through the future of retail technology and spearheading the digital transformation of the supply chain space.

Prior to her current position, Maria held CTO roles at various consumer technology companies including Omada Health and One Kings Lane, and led technology and product teams at Peet’s Coffee and Walmart.com.

Question 1: First Job: You’ve had a very technical background for a long time. It’s awesome to have a female leader in technology. Could you talk a little bit about your background and how you got to the position where you are now?

I’ve actually had quite a meandering career. But I like to think about it as all of the stars aligning for this specific purpose in life. I’ve been in large companies like Walmart, and really small Series B companies, with most of my focus around some sort of consumer experience.

I’ve also ventured out into healthcare. And now working for Albertsons, I feel like this is all coming together. There are both retail and healthcare aspects in my current role, within wellness and the pharmacy. And it’s really about trying to create better value, improve consumer lives, and develop relationships with the customer through the technology that we leverage.

It’s really exciting to be where we are, where the whole industry is being redefined.

Question 2: Generative AI: Everybody’s talking about it. But you, as a business leader, have seen other cycles before. Could you put a little context around Generative AI? Is this the highest priority for you right now?

I’ve seen these different innovation hype curves, but this one feels different for me. And the reason it feels that way is that it’s not just a new technology, it has the potential to fundamentally transform how we work (as opposed to just doing things faster). And that’s why I find it both exciting intellectually, but also transformative from the business perspective.

It can create or amplify the business models that we have right now and unlock new opportunities and capabilities for us. So it’s something that’s a very high priority.

And, as always, when you have something new, you also want to be careful about how you introduce it into the workplace, and the enterprise. You want to be responsible about it. We have this joke about Generative AI: “With great power comes great responsibility.” It’s a very powerful technology with a very powerful capability that must be governed appropriately. So there’s a lot to think about, and a lot to do. And then it’s: How do you scale it? How do you make it a fundamental piece of your enterprise business?

Maybe you can unpack that for us. I hear a lot of CIOs are doing experiments internally, but at the same time are focused on employee education. Could you talk about some early learnings?

There are a few different areas we need to focus on: first, investment is really hot right now. We’ll need to invest ourselves, and also become better and smarter through partnerships with other companies that have made their own smart investments. When you think about all the copilot capabilities, I do believe this will help enterprises to be much more effective.

And then finally, you need to think about the solutions that you create as a development organization. These will likely be more unique, more bespoke. Something that has specific value at a given company, and every company will be different.

Generative AI is still new and it’s already wonderful at certain things (while still learning others). There are going to be more “Aha” moments and early findings. So it’s key to get started where it’s strong, and not just look at it as a magical unicorn. Gen AI is going to be able to solve certain things, but not others, just yet. But as time goes by, more and more opportunities will become available.

So we have to think about being deliberate and thoughtful around what it’s already good at right now, what’s coming down the pipe, and what might be further out. Then, you try to figure out the phases of engagement.

The other optic here is thinking through quick wins and balancing those with impactful change. It’s not just prompt engineering, right? You also have to figure out the technology, the architecture, the security, the data governance…etc. You have to educate your teams. And so you want to both prove the value and also create a space within the use case to do all those things.

And you want to make sure that you pick the right use cases and the areas that have the highest feasibility. Do you have the data to make it happen? Can you really stand it up and get it off the ground?

And at the same time, you also want to start thinking about the areas that will provide the greatest impact: something that’s meaningful change. That will probably take longer.

Question 3: New metrics: You and I talked a little bit about your way of measuring impact or metrics. I’m curious, as you think about planning for next year, there are lots of use cases where Generative AI could be used. How will you go about deciding where to put attention? I love the idea of quick wins, but ultimately you said this could be quite transformative. So how do you balance all of that?

We try to evaluate two things: One is productivity. Is this something that will help us? Will it amplify what we’re doing as an organization? Will it help us achieve something in an easier and or faster way?

Those initiatives very often fall in the quick wins situations, by the nature of the fact that they give you productivity. It’s something that’s supposed to give you results faster.

And then you always want to pick something more strategic. And yes, it will take longer, and needs to be evaluated against the long-term impact.

You have this opportunity to start building things. I presume you’ve got new use cases. But then we’ve been in an environment for the past decade or so where we’ve been buying solutions versus building. And now we’re in a struggle where you may have to build things that are unique to your use cases. What are some of the technology gaps or issues you need to address? Where would you like more investment? On technology? Or is it in people and processes? And how do you balance the traditional ‘Buy’ versus ‘Build’ equation in this new sector of AI?

It’s interesting, I actually don’t look at it as a struggle. To me, it’s an extension of the same thought process. When you identify something that provides a utility to your organization, it makes sense to see if someone else is doing it and doing it well. In that case, you can buy it, and then you wind up seeing the results and benefitting very quickly.

But if you have things related to your IP, or unique to your knowledge or business model, you may not find something readily available in the market as a generic solution. In that case you’re going to have to build.

It’s the same logic that a lot of companies apply to the regular buy versus build.

In order to build, you will need to acquire more skills. You will need to train your team on this and find ways to be proficient. In some cases, you may be able to shoot for a hybrid solution that extends some off the shelf models that are already close. So maybe you don’t build things from scratch, but it’s also not a SaaS product right away. You’re able to take a platform and extend that platform, making it a little bit more unique, and a better fit for your needs.

Question 4: What are some of the issues or roadblocks, or maybe even gaps that you believe need to be solved before you can fully execute on this? What are the headwinds you’re facing?

I don’t look at them as roadblocks in the sense that we have to solve all of them before we can make progress. To me, they are those forces that sometimes play against you, and other times you just have to keep in mind while making progress towards your goal.

As this is a new discipline, some of these roadblocks will be new and unique, or even use case by use case. So as an example, when I think about the financial aspect of being able to educate people on FinOps, it would make a big difference in certain use cases, but in others may not be necessary at all.

The same applies to data quality and availability: depending on the use case that you’re trying to address, it may be a really big roadblock, or it may not matter at all. That’s part of the way you want to try to evaluate. Do I have all the things that I need to make my use case successful?

And then you need to think about evaluation and validation. How do you know that your use case will produce results you’re willing to stand by? And how are you planning to approach it? This is the whole idea and the whole discipline of data governance. You have to get this right from the very beginning and you have to understand your point of view. What are your guardrails, ethics, and social responsibility? A lot of education needs to happen, and the regulatory side of the house is constantly evolving. But this part is important, new, and ill-defined.

Finally, you must educate your team and upskill them in a variety of ways. It can feel daunting because it’s so new, so figuring out ways to be practical is probably the most important part. People may be excited, but they also may be scared, so you must make sure that you educate, create awareness, train, and figure out how to reach all of the right audiences in the manner that would be important to them.

How do you think about responsible AI as an IT leader? Do you educate your board? How does Responsible AI come together in your mind?

I don’t think it’s one person’s job to figure this out. I do believe it’s important enough that leaders of different disciplines within a company will come together and make it their priority. And so, depending on the structure of the company, it could be different people. But I do feel that this is one aspect that must be addressed early, and people have to be really thoughtful about it in executive leadership.

Maria Latushkin is a technology and product executive specialized in Retail, Payments, and Supply Chain. She is GVP, head of Technology and Engineering at Albertsons, a famous grocery short chain all around California, and the West Coast, where she is responsible for all aspects of technology for Retail, Payments, and Supply Chain, thinking through the future of retail technology and spearheading the digital transformation of supply chain space.

She is a high-energy, innovative technology executive accomplished at aligning technology strategy with business goals and delivering new business capabilities through technology.
Maria has spent over 20 years leading Product, Engineering, and Operations teams transforming technology for global brands including Walmart, Peet’s Coffee, and most recently Albertsons.

Before her current position, Maria held CTO roles at various consumer technology companies in the consumer and healthcare space, including Omada Health and One Kings Lane, and led technology and product teams at Peet’s Coffee and Walmart.com


You May Also Enjoy