Solving Hard Industrial Problems with Fast AI Deployment - with Kriti Sharma of IFS Nexus Black

Welcome, everyone, to the AI and Business Podcast. I'm Matthew DeMello, Editorial Director here at Emerge AI Research. Today's guest is Kreeti Sharma, CEO of IFS Nexus Black. IFS is a global enterprise software leader serving asset-intensive industries such as manufacturing, aerospace, energy and field service, helping organizations manage operations across complex distributed environments. Creedy joins us live from IFS's Industrial X event to break down how industrial leaders can move from theory to practice in AI adoption, not through long transformation cycles, but through rapid boots on the ground problem solving that delivers measurable operational outcomes in weeks, not years. Throughout today's episode, she explains how her team uncovers hidden data silos inside aging assets, turns domain expertise into production-grade AI tools, and partners with frontline technicians to solve real problems like batch loss, unplanned downtime, and workforce upskilling. Today's episode is part of a special series covering the Industrial X event from IFS, sponsored by IFS. But first, are you driving AI transformation at your organization? Or maybe you're guiding critical decisions on AI investment strategy or deployment. If so, the AI in Business podcast wants to hear from you. Each year, Emerge AI Research features hundreds of executive thought leaders, everyone from the CIO of Goldman Sachs to the head of AI at Raytheon and AI pioneers like Yoshua Bengio. With nearly a million annual listeners, AI in business is the go-to destination for enterprise leaders navigating real-world AI adoption. You don't need to be an engineer or a technical expert to be on the show if you're involved with AI implementation, decision-making, or strategy within your company, this is your opportunity to share your insights with a global audience of your peers. If you believe you can help other leaders move the needle on AI ROI, visit Emerge.com and fill out our Thought Leader submission form. That's Emerge.com and click on Be an Expert. You can also click the link in the description of today's show on your preferred podcast platform. That's Emerge.com slash ExpertOne. Again, that's emerj.com slash expert1. Without further ado, here's our conversation with Critty. Critty, welcome to the program. It's a great pleasure having you. I'm very excited to be here with you. Absolutely. This is my first time on the show hosting in person rather than on a Zoom call or anything else. And it's been a great pleasure to be here at Industrial X. It's been a really, really insightful event so far. You talked about really the importance of moving from experience to results. And we hear that all the time from industrial leaders on the show. The challenges around aging infrastructure, supply chains, data silos, and that this all limits agility. All at the same time, competition is demanding faster transformation. It's all coming to a head. You described a next phase in moving from theory to practice and how this means organizations can operationalize intelligence just across a very complex set of assets in an organization. If you can summarize for our listeners this part of your presentation today, what does this look like out there in the field based on your observation? So my team and I work in the industrial environments day in, day out. What that means is we go boots on ground. Say we don't just build our technology or our products in a vacuum. In this really clean model of going to the factories, the plants, the sites, aircraft hangers, distilleries, you name it. You call it a chemical plant. That's what it is. Exactly. We spend our time on the ground and really understand what works and what doesn't work. And when you see things with that proximity to the people who are really going to use the technology, you end up making things that they actually want. And so when we talk about industrial AI and the passion that we have for solving these problems, for me, it comes from being on the ground and building the relationships and doing it really, really fast. Right, right. You talked a lot about what Nexus Black does with these different clients. Give us a little bit of transparency into what you see as the mission for Nexus Black within IFS's AI operations. We solve the hardest problems for our clients, and we deliver value and outcomes in three weeks. Wow. And I know that sounds outrageously short amount of time. There might be a lot of skeptical reactions to that statement Hey count me Count me one of the skeptics because I worked in AI my whole career and I seen the different variations of it in various environments The reason why I say that it is coming from hard truth and facts, as I mentioned, our model is not just we build a product in a vacuum and chuck it over the line and then hope for the best or go into a two-year implementation cycle so you're still waiting for the results and, you know, life is just life. The model we operate is we go boots on ground, deeply understand the problem, rapidly develop products. We use AI ourselves to build it in a production grade way. And so within a few weeks, you start to see as a customer value and outcomes and delivered results. Now, don't take my word for it, right? I'm a builder of these things and our business is about AI and building in new products. But really, it's about the results that our customers are speaking of on our behalf. And you heard some of those stories today. Definitely heard the Hendrix distillery story. That surprised me. And I know we were talking a little bit before we turned the microphones on about that three-week window, especially to really get a relationship with the SMEs. Just take us through that process a little bit. And even where you have that window of three weeks, even as a goal to kind of be in and out and make this a fast transformation. What are you bringing to the table just in terms of really understanding their pain points? So we can deliver these outcomes and results fast because we bring in a few things with us. First, it's subject matter expertise. The reason why we're housed inside of IFS is to be deep domain expertise. Now, I would say that again. Skeptics on this call would probably think, oh, well, everyone comes here and they talk about subject matter experts and I haven't seen the results. And I get you. we turn that expertise into something quite special we combine it with the world's best AI engineers it's an elite team it's a you know the smallest and you heard anthropics say today their reaction on the quality of our team so we we bring the two together and let me give you an example of what I mean by that by bringing expertise into AI and build turning it into products at the distillery this is William Grants and Sons they are the brand behind Glenn Fittig-Whiskey and Hendrix Gin. We spent time on the ground. We understood, oh, real data and information is sitting in their piping and instrumentation diagrams. It's sitting in the SCADA systems. It is sitting in historical work orders of maintenance effort. Now, our AI platforms and capabilities are so sophisticated and trained to do exactly that. We understand how our AI tools and capabilities understand how to read a complex piping and instrumentation diagram. Right. How did it learn to do that? You bring together industry expertise and you combine it with AI capabilities and you build next to your clients over and over and over again. You productize, you scale, you do the hard work. We're not doing it our first time. Yeah, absolutely. You were saying before, I think we caught it during our recording, but you saw them as a chemical plant. And that makes, of course, a lot of sense. But I think especially considering them as a distillery tends to nail what their aging assets are like. And that would, I would assume, be where you're tailoring the solution. Is it more just about finding where the data is for the client and building out from there because you have that subject matter expertise? Or how much are you tailoring it based on the aging assets, who they are as a client? It's finding their problems and how we solve them. In this case, the problem was very clear. It was an issue of production batch losses. Downtime turns into less product being created. So in that case, we saw 38% of the repair work that they did was emergency or corrective. That was a big opportunity. That means downtime that you could slash. If you're planning the work better, if you're predicting it, you're preempting it. now people would say wow i'm like i'm my own skeptic here i get it i i'm not a you know ai washer this is not what i do right i don't even want to use those words don't want to use the words ai if i can avoid it um but really this is about um how we solve the same old problems but in entirely new ways right now with the with the versions of technology and how it has evolved so fast we can rapidly create tools that connect into multiple different data systems there are new new techniques to do that we can understand and interpret data from many different modalities it not just text that ai is able to interpret we can do a lot of advanced work in sensor data pressure temperature vibration sound images video P diagrams Absolutely You really leaned into the Hendrix Distillery use case in today presentation Any other examples maybe outside the chemical space that kind of demonstrate that facility of staying within the three-week window and building that proximity to such a parameter? I'll give you an example of windscreens that we talked about today. Disclosure, I don't drive. I live in London. We take public transport everywhere. The story we talked about today is one of windscreens. In the past, it used to be as simple as replacing a pane of glass. Right. Not anymore. Most cars these days are fitted with advanced driver-assisted systems. This means the kind of thing you see when your car is telling you to stay in your lane or you're about to spot a hazard or when to start to slow down and apply your brakes before you know you need to. These sensors are often connected or calibrated to the windscreen. Now imagine a technician out in the field who has now historically done replacement of the windscreens and now you have to do something as complex as literally a very smart computer that's connected to the windscreen. And these are getting more and more complex, many different kind of vehicles coming to turn. So the challenge with Auto Windscreen was, well, how do we get new talent or old talent upskilled into solving much harder problems than they used to while continuing to do more jobs per day and maintaining the margins? I saw that as the buried lead really through the whole media coverage of AI as I kind of see it, especially where I've expressed shock on the show. And I think it's an assumption, maybe a lot of our listeners have, that manufacturing can be departed from tech. But I think a big reason that's not true anymore is the Internet of Things. You're starting to see more and more technology in this age of AI in mundane objects like windshields, coffee cups, whatever it is. Just wondering where you're seeing that across the board, maybe how industrial leaders can best prepare for that future where they're just not making coffee cups anymore. They're making smart coffee cups anymore. And we're not just trying to fix mundane items. We're trying to keep them connected to a larger data mainframe. I would think about hard problems that haven't been solved before and go at them. but don't do it for we're going to run a project and we're going to spend two years on it and see how much what comes out of it i do think there's a tremendous amount of value to be gained by going after specific use cases that may be hard and that's the value now you know we don't this is not the moment for small incremental steps or building yet another chatbot that you can ask a question to and get an answer and people will sometimes use it and sometimes not now is the time to challenge the partners who come and want to work with you to say how can i drive meaningful impact on my bottom line now in the trade-off there that i find or at least for our listeners where they might struggle is i got to find something safe enough where i know i can sandbox it i know i can keep it safe but it is mission critical enough that it matters and i want this project to succeed it's not just this little experiment on the side where great if it works out if not i move on with my life any advice on just trying to strike that balance or maybe a criteria by which they can think of those problems in maybe find a category where oh that's easy to sandbox i think it's starting um so thinking about these problems but then looking at how you can increase the autonomy over time so to give you an example could be production planning it remains a hard problems to solve is still unsolved to a large extent. How do you take all the constraints into account and all the chaos that happens last minute of a rushed order has come in and you've got to figure stuff out. If somebody has gone off sick, you know, this is not the stuff of theory. Or a repair has, emergency has sent the whole plan into whack. And when you deal with that sort of world, it might feel, how can an AI solution really help here and completely automating it? Well, the goal here is to really drive higher throughput and do better planning. Now, you could start to dial up the autonomy levels over time. So you might want to start with, well, that's the problem we want to solve, but there might be more intervention. There might be more controlled deployments. But over time, as you see more success, you can just give the tools more agency to take actions on your behalf. And the value here in starting in this way, keep the eye on the big prize but take small steps forward I often talk about it with my team as our head is in the cloud our feet are firmly in the mud Right. When you think hard about the current problems, you dream big. And the way we solve it is you think about the big problem, solve the big problem, but do it in small steps. As we often hear on the show, it doesn't even need to be that you solve 100% of the big problem. It's really just maybe you can move the needle 10% on a big problem that is important. And that's enough of a carrot to really get you by in the short term. Any other takeaways you're hoping leaders in the audience today across the industrial space, even listening right now, will take from industrial? Don't listen to me. Listen to what my customers are saying. And I cannot emphasize it enough. I love building products, businesses, teams, solving problems. I believe the best way to drive impact is let others do the talking on my behalf. So today you heard those stories. Our partners, our early customers, they're saying those themselves. Listen to your peers. Yeah, I think there's a lot of really great advice in the community as well. but also if you can still get out of the process that you've got out of, you know, making the robots in high school, I have my, I have my equivalent of that recording bands and stuff, but yeah, it's, it's, um, if you can still get that, that, that same result for yourself out of that, I think, I think that really drives customer results. Um, really appreciate you being with us on today's show. Thanks so much. Thank you so much. Wrapping up today's episode, I think there were at least three critical takeaways for industrial operations leaders, asset intensive manufacturers and executives overseeing data and AI strategy in complex physical environments from our conversation today with Kreeti Sharma, CEO of IFS Nexus Black. First, rapid AI impact in industrial settings is possible when teams start with frontline reality, not theory. Embedding with technicians, understanding where data actually lives inside aging assets, and treating domain expertise as a design input dramatically shortens time to value. Second, solving high-value operational problems does not require full autonomy on day one. Leaders can begin with constrained, safe-to-sandbox use cases like predictive maintenance and production planning and dial-up autonomy only as confidence, accuracy, and ROI grow. Finally, meaningful transformation comes from tackling hard, mission-critical workflows rather than incremental experiments. Even moving the needle 10% on downtime, throughput, or technician efficiency can deliver outsized returns across an entire industrial environment. Are you driving AI transformation at your organization, or maybe you're guiding critical decisions on AI investments, strategy, or deployment? If so, the AI in Business podcast wants to hear from you. Each year, Emerge AI Research features hundreds of executive thought leaders, everyone from the CIO of Goldman Sachs to the head of AI at Raytheon and AI pioneers like Yoshua Bengio. With nearly a million annual listeners, AI in Business is the go-to destination for enterprise leaders navigating real-world AI adoption. You don't need to be an engineer or a technical expert to be on the show. If you're involved in AI implementation, decision-making, or strategy within your company, this is your opportunity to share your insights with a global audience of your peers. If you believe you can help other leaders move the needle on AI ROI, visit Emerge.com and fill out our Thought Leader submission form. That's Emerge.com and click on Be An Expert. You can also click the link in the description of today's show on your preferred podcast platform. That's Emerge.com slash expert one. Again, that's Emerge.com slash expert one. We look forward to featuring your story. If you enjoyed or benefited from the insights of today's episode, consider leaving us a review on Apple Podcasts and let us know what you learned, found helpful, or just like most about the show. Also, don't forget to follow us on X, formerly known as Twitter at Emerge, and that's spelled again, E-M-E-R-J, as well as our LinkedIn page. I'm your host, at least for today, Matthew DeMello, Editorial Director here at Emerge AI Research. On behalf of Daniel Fagella, our CEO and head of research, as well as the rest of the team here at Emerge, thanks so much for joining us today, and we'll catch you next time on the AI in Business Podcast. Thank you.