In this podcast episode, Dr. Beverly Wright, Vice President of Data Science & AI at Wavicle Data Solutions, speaks with Iddo Salton, Vice President of Innovation, Data and AI at Delek US. Together, they discuss navigating cross-industry moves in data science and AI, exploring the challenges and opportunities of shifting from satellite and space technology to oil and gas.
Speaker details:
Watch the full podcast here or keep scrolling to read a transcript of the discussion between Beverly and Iddo:
Beverly: Hello, and welcome to Tag Data Talk. With us today, we have Iddo Salton, Vice President of Innovation, Data and AI, from Delek US and we’re talking about navigating cross-industry moves in data science and AI. Thanks for being with us, Iddo.
Iddo: Thank you for having me. This is really exciting.
Beverly: I’m really excited to talk about this subject because there’s so much debate over this side of data science and AI, especially the talent, like the talent thinking about what is it like to go across industries? Should you stay in one industry? Can you move across industries? So, we’re going to jump right into it. And I want to start off with a little question about your background. Tell us, why are you so cool?
Iddo: Wow, maybe I’m cool, I don’t know. Because of working with data and AI.
Beverly: There you go.
Iddo: Maybe it’s a good reason, yeah. And in the last few years, I saw a lot of changes and shifts in this world. And actually, I also had the transition from one industry to another. So that’s definitely provided me a good perspective on the data world and these kinds of technological aspects.
Beverly: Yes, yes, I love it. And where did you go to school? Like where’d you go to school to study?
Iddo: So I was born and raised in Israel. I started my career over there. I studied in university, both 1st and 2nd degree in electrical engineering and most of my career I was in the satellite and space industry.
Beverly: Oh nice.
Iddo: Yeah. And this is very exciting. And almost five years ago, I moved to the oil and gas industry. I moved there with my family here to the US.
Beverly: Was that in Texas?
Iddo: Nashville area, Tennessee. And learning a new area is very exciting and providing knowledge to learn these kind of things.
Beverly: Love it. So tell us, what does Delek US do? Like some people don’t know what that is.
Iddo: Sure. So Delek US is an energy company, whatever is considered to be a downstream and midstream company. We have 4 refineries in the US, two in Texas, one in Louisiana, one in Arkansas. And we have also the midstream segment, which contains pipelines, tracking rail, how to transfer the crude from the origin of the crude till the refineries and from the refineries to do the refining and then transmit the products to those that sell these their products. So that’s the midstream segment.
Beverly: OK, it’s funny. One of the longest I’ve ever been anywhere was in the utility world. So, I worked for Southern Company, Atlanta Gaslight, and I always found it really fascinating. I don’t know, you know, utilities kind of get a bad rap, but it’s super fun, right? So we’re talking about navigating cross industry moves in data science and AI. So what drew you to the idea that like, oh, I don’t have to stay in this one sector, I can kind of move around. Was there some kind of light bulb that went off in your mind, or what was your philosophy behind that?
Iddo: So it’s a great question. I can tell you that the opportunity to explore new world is something that fascinated me. And in the space and satellite industry, I was very well familiar. Of course, every day I learned new things, but when I had this kind of opportunity to go into a world that before then I heard the term crude oil and suddenly I had the opportunity to learn some more, to learn exactly how this industry works or, or it’s part of the industry. For me, it was fascinated, and I wanted also to test to go outside of my comfort zone to see how how I’m able to learn a new topic, learn a new industry, learn new business and that that was the main reason for me to to go into this.
Beverly: You wanted to kind of, you wanted to stay with what you were doing, but you wanted a change. And so you sort of chose to change the surroundings and kind of keep your core of what you were actually operating on, you know, the data side of it. That’s a really neat idea.
Iddo: I was an engineer, so I didn’t plan to work in a refinery or to be an operator. I knew that these kind of skills you need to work all your life in order to gain it. I wanted to focus on the data and AI and innovation and technology and to bring value to to the company.
Beverly: Yeah, this can be a really intimidating thing for a lot of people. They may feel very uncomfortable with. Like when I started, started a long time ago and when I started, there were, you know, 6 analyst jobs in all of Atlanta. So I didn’t really have a whole lot of choice because it wasn’t as popular as it is today. If I were to restart, I would probably pick an industry, and I always loved financial services. So I’d probably pick financial services or the nonprofit world and just stick in there. So when you decided to shift, what were some of the things that you were like, whoa, that is crazy different. Were any of them related to the core function of data science and AI? Was it all sort of outside of that realm?
Iddo: I think eventually it’s all about the core business of each industry or segment. And I can tell you that one thing that I learned about the oil and gas industry, the core business is the chemical engineering is the foundation is the refining processes. And when you come from a world that is full and based on technology like I came from on the satellite and space, this is a very digital-heavy environment. When you move from this kind of environment to an oil and gas industry which is very conservative and traditional, you know the refining process was established more than 100 years ago and still the core business is the same. How you take care, how you get the crude oil, break it into hydrocarbon products, and then sell it. And the technology component is only an enabler here. We can definitely talk about the data, and data brings a lot of value over there, but eventually, the core business is based on chemical engineering, which is the foundation.
Beverly: That’s so bizarre. So, does that change how the data is seen? Because I’m just guessing here because I’ve done some transferring across industries as well. And in some places where, you know, chemical engineering might be the king of that company. You look at the COO or the CIO or the C, whatever, and the whole C-Suite, it’s all chemical engineers, you know, did you see that kind of phenomenon? And does that change the nature of the type of data science and AI kind of work that you can accomplish?
Iddo: So, yes, because you come to a company that you know that the data science is not the core, but I must say that with our management and decisions that were made by our executives, they decided to go into more innovation, place more of data science. We established the team, our CIO, the current CIO joined to the company and established a lot of efforts to bring the data and to be much more decision enabler. And I think that we are in a very good pace with regard to that. Still the core business is refining, but it doesn’t mean that all the envelope processes, all the finance, all whatever we’re doing to the digital enablement of the work that is done in the refineries, it doesn’t mean that we are not able to improve that.
And another effort that we made, we brought new technologies to support the operations in the refineries. Our first core value is safe and reliable operations. We need to make sure that our people come in the morning to the refineries, to the work, and go back safely to their families. And we want to bring technologies to enable that, to reduce the risk and to improve the safe environment. And we found good technologies and and tools that we were able to do that. And that’s something that is technology provider as innovation, as data, we are able to support that to the business.
Beverly: Now, do you think that it was maybe a little less friction filled because of your experience with satellite and digital? Like do you think that it’s actually beneficial for different industries to sort of blend and, and information share and that sort of thing? Or is it better if you get, you know, people that are dug deep into that one industry?
Iddo: So when I joined, obviously it was very difficult. I’m coming from one industry to a new industry that I’m not familiar with. With time, I found out that the different perspective brings value. And I can tell you that one thing that I knew but also is something that I recommend, I knew that eventually the business, the core business is, is the key. So, working together with the people on the refineries, working together with our operators, working together with the people in the different business units.
Even if you think that you know everything, especially if you come from a data world and you think that data in one world equals necessarily data in other world, I think you need to reconsider because every data eventually is connected to a business problem. And it all starts with the problem and with the business. So I think this is something that helped me to bring my perspective to my role.
Beverly: So you brought up the the phrase. The phrase is ‘is data data or is it conditional?’ You know, is data somewhat conditional or is the analytics, the data science or the AI that you apply, is it somewhat conditional based on the the type of industry that you’re dealing with? So what would you say to somebody that said, you know, when I’m working with healthcare or I’m working with financial services, data is data. What would you say to that?
Iddo: I say that there is a lot of truth in it, but I think that the foundation or the core to it is being related to the business problem. Eventually data is data, but if it doesn’t have any context, if it doesn’t relate to the business problem, it means much less if not at all. So, my recommendation to this kind of thoughts, make sure that you know what is the business around and then the data will will support it.
Beverly: So focus on the business problem first. Not like, oh, I know how to do this cool thing, or I can apply this neat technology. If you start with that, you’re going to trip up.
Iddo: Exactly. And I want maybe to classify 3 areas. The first one in such kind of industry that data is enabler. So, for sure the former statement is valid. The second classification is in areas where data supports decision. This is also something that is very relevant to the business. But there is another class, the third class that the data is the product. So, so in this kind of areas where data is product, maybe being more expert on the data itself is a big advantage.
Beverly: And the industry is less important, so.
Iddo: Still important, but there’s 3 levels for sure. The first two, it’s very important, right? The third one is also important, but but a bit less.
Beverly: Yeah. So not everybody can do this by the way. Not everybody can, you know, swim from one pool over to another. They sort of feel comfortable in financial services or in healthcare, whatever, because the regulation is different, the conferences alone are different, the learning is different, the maturity level is different. Like there’s so many differences. So, what are some of the things that people tend to kind of mess up if they move from one industry to another? Like if you’ve seen other people come in to say your work or a prior work life or whatever, and they came from another industry, what are the certain things that you tend to see, like, oh man, they’re thinking about, you know, they’re thinking about this wrong.
Iddo: I think that when people try to mimic exactly what they did in their former life to the new life, it’s not going to work unless you understand the business need, you understand what is the exact use case, even if it’s the same use case to your previous company, for example, predictive maintenance, budget estimation or forecasting.
Beverly: Yeah, yeah, of course.
Iddo: So even if there are some different use cases or similar use cases, but from different industries, still everything is with the context to the specific business.
Beverly: Even predictive maintenance, like oh man, I got this down. Predictive maintenance and utility versus predictive maintenance and automotive even. So, it sounds like you’re saying, don’t go in with these ideas of just like, oh, I already know how to do it. It doesn’t matter. You have to be a little bit of a sponge, right? And sort of learn the environment and the things that are core to that business and to the function first.
Iddo: When sometimes, people come with previous experience, they in many cases come with a solution before hearing and understanding what the problem is. When we hear the problem, when we hear the voice of the customer, what the customer needs, we will be able to provide much better solution versus to try to solve it in a way that worked in the previous use case or company. So that’s something that I definitely recommend.
Beverly: OK, so don’t come in with like, here’s the answer. Now let me get back into what the question must have been. You know, we’re assuming that you understand the original question. Like my team and I did some safety studies, and one was with a transportation company, one was with manufacturing, and one was with a retailer. And they were completely different. I mean, the funny thing too is it wasn’t even the same role of person asking for the safety study. You know, they were all wanting to reduce safety incidents, but it was different people. One was a CFO, one was a safety manager, and I forget what the third one was like, maybe a chief HR officer. So it was interesting to see how that kind of played out. And I think you’re saying something similar, that don’t come in with like, here’s our package solution and this is what we think needs to be done without really fully understanding and immersing yourself. And that’s probably the main place that people trip up. What about the actual data, like the OR the maturity level of the organization? How much impact do those have?
Iddo: It has a huge impact when the organization is mature for data. This is something that, in the culture perspective, the organization understands the data can bring value. Luckily, we are in a very good spot from that perspective. The people in our company thrive on data. They want more and more data. They understand that the data, they can trust on it and then make a decision. And by the way, that’s also a journey because sometimes you have data that you can’t rely on. And I think that one of the areas today that is very important is data integrity.
You need to make sure that whatever you put in and consider data, you can rely on it. And if you find some kind of an error or mistake in the data, you, you’re not going to trust it. But if you see that you can trust it, you can make a good decision, and the maturity level of the organization is very important from that perspective. And even if by the way, every organization has maturity level, still it doesn’t mean that there is no room for improvement. But as long as the organization understands that the maturity level is X and you want to go to double X or to improve it and you invest the efforts, I think that’s the right approach.
Beverly: So, in thinking about this navigation across industries, are there some industries where you say, like, oh, no, you can’t do that. You can’t go from, you know, this industry to that industry. I didn’t mean that far. Are there somewhere it’s more seamless? And are there others that are just like impossible?
Iddo: So, after moving from the satellites to the oil and gas, I wouldn’t say that there is some kind of no-go.
Beverly: There’s not a no-fly zone or anything like, you think you could go from government to retail? I mean, that’s what I think about, is I think about something like government, where it’s very different. And then I think about retail and CPG, which is arguably one of the more advanced. I mean, maybe healthcare technology is starting to get there, but retail CPG is hyper-personalizing. And so, they’re sort of, you know, kind of last time I checked, leading the pack and some of the AI implementation. But you think even going from, and that’s another question too, is, can you move up in the maturity, or do you always have to move down in maturity? Like, where do you see all that?
Iddo: First of all, I think that data is data from that perspective. So even if you have a data role, AI role in specific industry, 80 percent at least will serve you well if you take it to the next journey. Yeah, the other 20% which are the core. It’s not the major number, but it’s the core, it’s the foundation. It’s knowing the business. So, I think that as long as you are humble enough to know that, OK, I know about data, but I don’t know about the business. I need to learn, I need to work with SMEs, I need to work with the business owners.
As long as you understand that, I’m sure that you will be able to provide good solutions after you hear the problems. And that’s something that I really believe in. And to your second point, I think that the maturity level is something very important. Obviously it is recommended to go up with the maturity level, but even if you move from one company to the other and the maturity level is a bit lower, OK, you know what good looks like. So, you can definitely help the organization bring them up.
Beverly: Right, right.
Iddo: I think that’s also a good challenge from that perspective.
Beverly: Yeah. And how can someone who’s moving across industries and they really need to immerse themselves and better understand the industry and what leads it, and how it operates. What’s the best way to do that? I know you’re talking about working with a business, but what does that look like?
Iddo: Learn a lot. For example I can say that when I moved to the oil and gas industry, I didn’t know about the refining process. So, I try to learn it. For sure, I won’t be a person that worked in this industry for years, but at least I try to learn the foundation, ask the questions and start to understand, to better understand what the gaps are, what the needs are. And I think that collaboration, in today’s world, there is no one person or one organization that holds the solution. It’s all about collaboration and working together with the people. And with all due respect to data and AI, which I really respect, eventually it’s all about the people. It’s working with the team.
Beverly: That’s a great quote I love.
Iddo: Collaboration.
Beverly: So regardless of all that, it really gets down to the people. I get it. Certain last couple of questions here. If someone’s considering like taking a dramatic move, you know, and they’re saying like, gosh, I don’t know, I’m so tired of telecom or I’m so tired of retail, you know, or whatever. If they’re really thinking about making a jump, are there certain characteristics or attributes or mindsets of the employees or different units that would say, oh, you’re going to be fine. And then someone else may be like they’re not going to be fine. What makes someone a good cross-industry transition person?
Iddo: I would mention one main characteristic and it’s curiosity. By the way, our company, Delek US, selected curiosity as one of its core values. I really think that curiosity, if you’re curious enough, or I would say differently, you must be curious enough in order to move from one industry to one industry to learn a new topic, to know how to bring value to your organization, especially in technology. Yeah, few years ago the pace was much lower than today, right?
Beverly: So fast right now.
Iddo: Today the pace is amazing. It’s like an exponential pace and and you need to be curious in order to meet the pace. I would say maybe from the other end, just to balance it, you also need to stay focused. So, because if you lose focus, there are so many things out there that you won’t know what to focus about, right. On one end, you must be curious and to pick up the right level of focus. But curiosity definitely is key for success.
Beverly: Oh, I love that. And I love that your employer chose a highly human characteristic for one of their core values. That’s great. So to end off, what final piece of advice would you give to somebody who’s considering cross-industry moves in data science and AI?
Iddo: I would say, first of all, I go for it. I think it’s a great challenge and a very exciting move. So definitely, I would encourage it, yeah. And make sure that you understand that there is difference between each and every industry.
Beverly: Going humble?
Iddo: Going humble, yeah, the data piece might stay very similar, right? But it’s all about connecting it to the right context, to the business problem, to the industry. And I think that what makes this exciting and also the data and AI world makes it so exciting because it’s related to all kinds of industries.
Beverly: Yes, I love it. Thank you so much to Iddo Salton, Vice President of Innovation, Data and AI at Delek US for talking to us about navigating cross-industry moves and data science and AI.
Iddo: Thank you.
Cross-industry moves in data and AI demand both curiosity and humility. While core data skills transfer across sectors, true impact comes from learning the business context, aligning with industry needs, and collaborating with people closest to the work. For professionals willing to step out of their comfort zones, such transitions can be both transformative and rewarding. Explore the full catalog of TAG Data Talk conversations here: TAG Data Talk with Dr. Beverly Wright – TAG Online.