What's Up with Tech?

From Silos To Systems: Cross-Enterprise AI With R4

• Evan Kirstel

Share episode

Interested in being a guest? Email us at admin@evankirstel.com

What if your enterprise could make the right decision, across every silo, in real time? We sit down with Paul Breitenbach, Founder & CEO r4 Technologies, to unpack how predictive AI goes beyond generative tools and actually drives outcomes: higher revenue, lower costs, and fewer bad events across the board. Paul explains why cross-enterprise management matters now, how it automatically pulls data from legacy systems, and how it sends decisions back without ripping anything out.

We walk through tangible examples that make the strategy real. In supply chain, predictive logistics ensure the right product hits the right shelf at the right time and price. In security operations, microlocation and environmental signals re-route guards on the fly, cutting incidents by 50 percent. And in a powerful “tech for good” case, R4’s Smart Food program identifies surplus food and matches it to SNAP demand, effectively doubling or tripling buying power while reducing waste—turning a national problem into a solvable coordination challenge.

Trust and adoption come from transparency and speed. Paul shows how confidence scores, explainability, and human-in-the-loop controls help leaders move from gut feel to measurable, auditable decisions in days, not quarters. We also explore what’s next: a more intuitive V5 interface, tackling higher-complexity domains like healthcare and energy transition, and why 2025 could be the tipping point where predictive AI becomes the enterprise standard.

If you enjoyed this conversation, subscribe, share it with a colleague wrestling with silos, and leave a review to help others find the show. What cross-enterprise challenge do you want AI to solve next?

Support the show

More at https://linktr.ee/EvanKirstel

SPEAKER_00:

Hey everybody. Fascinating chat today. We're diving into the world of enterprise results at scale with AI, a hot topic today with R4 technologies. Paul, how are you?

SPEAKER_01:

I'm great. How are you, Evan?

SPEAKER_00:

Doing well, really intrigued for this conversation. You've been at this for some time before the hype around Gen AI even came to the forefront. Before all that, maybe introduce yourself and the mission at R4.

SPEAKER_01:

Yeah, hey, it's great to be with you. I'm Paul Brainback, HEO, founder of R4. We're an AI company started by the founders of Priceline about 12 years ago. So we're um excited to be here. I feel like I've been in this data math thing for decades now. So it's like a new technology solving really important critical problems.

SPEAKER_00:

Fantastic. Well, Priceline is still my go-to favorite. So congratulations on building an amazing entity. But what problem does R4 solve today better than uh anyone else, in your opinion?

SPEAKER_01:

Yeah, R4 is uh a new AI company, really. When you think about um everyone talks about LLMs and generative AI, right? Um they're awesome. What R4 is, it's about using predictive capabilities of using data and math in real time to drive a prediction that really allows us to optimize things. And so we're known for supply chain, predictive logistics, talent management. Uh we've got uh great use cases in guarding and security, predicting negative things from happening and then preventing them, driving automated actions. But the new big idea in AI is you use data and math in real time to drive a predictive outcome that generates more revenues, lowers costs, uh generally speaking, makes things better, very different than an LM.

SPEAKER_00:

Indeed. And you're known for something called cross-enterprise management. Uh, what is it? Why is it such a breakthrough?

SPEAKER_01:

Yeah, well, all the systems that we've spent decades building and living in, right, where they're so siloed, right? So what happens is all the left and right hands of organizations aren't able to talk together. In new AI, with what we do, we can pull data together from all the silos all automatically. And even more importantly, the software now, our AI technology, can do that without the need for any data scientists. So think about this incredible, monstrous opportunity that if we could leverage all the data and make it actionable, all automatic. And I don't need just stabs, large staves of data scientists in order to deploy. Think about what a competitive advantage that is.

SPEAKER_00:

Yeah, I bet. And I I've worked in large enterprises like Oracle, you know, big data companies, and everything was siloed. So even within those kind of companies you think as big data experts. So how do you break down the silos? Because there's some part of it's cultural, part of it's organizational, but then part of it is uh data driven. How do you actually do it?

SPEAKER_01:

Yeah, well, with uh R4, with XEM or cross-management enterprise AI enterprise capability, the software pulls the data together automatically. And so what's so important is, and then we can configure it by use cases. So every silo can still have an outcome use case that it needs. So let's take a supply chain, right? In commercial world, right? There's someone that orders the product of your head, someone that puts it in distribution, someone has to maneuver it to a store, so to put it onto a store. There could be a dozen people, all with different silos, that operate, you know, as as hard as they can given current technology. Now, in the new world with new AI that R4 delivers, um, all of that data can be pulled together automatically to optimize exactly the right product, the right spot, at the right time, and the right price. And in fact, that's actually what R4 stands for. It was named by uh Igor Zucker, who was with us at the uh beginning of Price Sign. And it's actually R to the fourth. That means that in the new world, by operating cross-enterprise, revenues can go up while costs go down. It's really a transformation, it's the golden age of AI. So no longer are you just putting data into reports. Generative AI has been a really cool capability as well. Now you can literally drive automated actions uh by pulling data together automatically and deploying it without the need for any data scientists.

SPEAKER_00:

Amazing. And you're seeing big results across industries. You have some great examples on your website. Any stories or anecdotes about early wins that really proved out your model?

SPEAKER_01:

Yeah, I there's so many now between we have three lines of business commercial, defense and national security, and our civilian state, really with our our AI smart food program, and we'll talk about that. But like the the one of my favorite ones, I mean, like supply chain is cool, all the predictive logistics, super cool, meaning eliminating out of stocks, minimizing overstocks, pricing and promoting all properly, with you know, all this real incredible precision. Those are all super cool. They drive real revenue. But one of my favorite examples is a security guarding example, because it's one that you wouldn't think of. So in the security guarding business, typically it goes from point A to point B to C to D. And with R4, we pull real-time data together automatically together with microlocation data and environmental data, and we dynamically route security guards in real time. So from point H to S to B back to H to S, and it lowers negative incidences by 50%, five zero. So when you think about this, the promise of predictive capabilities like R4's XEM cross management system, yes, you can put things on the right shelf and the right quantity at the right price. That is super cool. But imagine the concept of eliminating bad things from happening. I mean, think about the addressable market of that, right? And then up until now, until you could see the entire organization and the the entire organization where that location would be, there's not been a way to actually lower, you know, to predictively lower uh and take it automated actions to lower negative incidents from happening. So that that's a favorite example. But we've got plenty across defense. Um, and um my other one is um our smart food program. So we're deploying the AI technology. Remember, price sign secret sauce was we we used technology to find excess inventory, right? That wasn't going to be sold, and we would match that to customer demand. Um it made made gave consumers an 80% discount, gave made industry you know, tens of billions of dollars of profit, right? And it created a monster company, number one company on the Nasdaq for quite a long time. Um R4, we're using the AI technology to find 30 to 40 percent of all the food we produce in the US is wasted. It's about 400 billion dollars worth of food.

SPEAKER_00:

Wow.

SPEAKER_01:

And so we're using the AI technology to find that to be wasted food and match it to the food stamp Snap customer, where that$300 that they're already receiving on their Snap card buys$600 and$1,000 worth of food. And so when you think about this, imagine with new AI never having negative things happening, and then imagining ending being so bodacious is to try to end the hunger problem that we have in the US. I think this is now, this is not only art of the possible, this is doable.

SPEAKER_00:

Wow, talk about tech for good. That sounds amazing. Uh, you mentioned making AI accessible, available to teams beyond the data science geeks and the spreadsheet jockeys we all know in the enterprise. How does that work exactly? How does it look on the ground?

SPEAKER_01:

Yeah, the user interface, you know, we're on V4 of our of XEM technology, and it's super cool. I mean, imagine super sophisticated math and technology deployments that the business user on the other side has no idea how the math works, right? And the new V5 that we're releasing shortly, yeah, it just even goes way beyond it. It's much more intuitive, much more attuned to the way a business user thinks, adding all of that predictive capability all in the background without me having to even know how to use a user interface. So I feel like, you know, people ask me all the time, it's like, hey, what should I study in school? I'm going to college, you know, we all have kids and uh grandkids going to school. And I'm like, you know, now that knowing what we're able to do at R4 and how new AI is going to change the game, putting this capabilities directly in the hands of business users, I'm like, I'm kind of big into humanities now, asking the right questions. Like, what is the outcome? What is the imaginative, you know, element that we could actually work on together? Because if you have a magic lamp, so now the trick is what to ask the lamp, right? When you have three wishes. So I think that's to me, you know, it's all about the user experience. And I think that's where our four and our cross-enterprise management technology just really hits it out of the park. Just easy to use by people who don't know anything about data science.

SPEAKER_00:

Amazing. Yeah, my son's a junior in college studying economics and statistics. And uh he always hated coding, never wanted to learn to code. He's coding apps now, you know, with these natural language models. So I was really surprised to see that even at his level. Um, when you talk about data integration projects and transformation in large enterprise, I mean, you measure these things in quarters, sometimes years, right? Um, how quickly can can you see value uh with these kind of projects?

SPEAKER_01:

Yeah, that's what that's a great question. I mean, there's kind of like in in our world here at R4 in the AI space, there's like a new Moore's law, right? So rather than being, you know, doubling capacity every year or two, you're now doubling capability in every week or two, right? In a month, right? And so gone are the days that we all grew up in where you spent six months a year writing requirements, you know, which you know, now you could do this in an hour and set and configure the technology for what outcome you're trying to get at the product. And this is at enormous scale. And because it's all human in the loop, we really believe in um, as you were talking about just a minute ago, human in the loop. And so because the humans are watching it, you're seeing what the machine is doing. And the um the speed of innovation now, I think it is like almost a moment of existentialness, right? Like if you aren't at the enterprise level using stuff like what we do at R4 or in at an enterprise scale, um, you're gonna be behind very quickly. And remember, there's a long tail to this. There's like blockbusters quite a long time to go fully away. But once it kicks in, it will be hard to ever catch up because now once all the employees know how to take advantage of these new predictive enterprise-wide capabilities, it becomes a the gold standard. So it's it's time to jump on. 2025 is going to look at as the year when it all began.

SPEAKER_00:

Amazing. And so you talk to C-suite all the time and executives. How do you get them to or help them kind of trust AI-driven recommendations when they've been relying on, you know, experience, maybe instinct over the decades?

SPEAKER_01:

Yeah, that's another really good question. The you know, the I think the fact that in all aspects, as R4 creates a prediction, it's telling you how confident it is that it's going to be correct. And it shows you why it's creating that prediction down to a very, very, very specific degree of precision. I think people learn that, like, uh, it's got it under control. So, and then I think that the trick now is the human saying, hey, tell me these things that you don't know about, that okay, show them to me, right? And that's the the balance between the human and machine collaboration that R4 lets happen. And it's really cool. In the beginning, it is like back remember in the days of you know, you were both old enough to remember the internet. Like, what do you mean I'm gonna buy things with a computer? What do you think I'm gonna put my credit card in this computer for? Like, how is that possible? Now it's like, what do you mean I need cash? Right. So I think the, you know, I I think as people start to experience it, they realize this is just a fundamentally better way to, you know, make, you know, drive productivity, whether it's whatever the job is, whether it's on the revenue side or on the cost side.

SPEAKER_00:

Fantastic. So I was in the enterprise tech world for, you know, as a practitioner for 25 years. And um, days could be there was a lot of grunt work, a lot of grind, a lot of systems you had to work through, brutal. And it's still that way in many enterprises. What are these changes that these these practices that you're driving enable for the average business user? Well, what what will it change on their day-to-day uh do you think the impact, not just on organizations, but on the individual?

SPEAKER_01:

Yeah, well, this is yeah, I think on the users, remember the most people really want to be good at their job, right? They they and the ungodly amount of work it takes to manually run them with spreadsheets and clipboards and and try to just make make the different systems all to you know work together, right? I think the the it's like a especially in the defense world that we're is like an eye-opening thing when they realize, oh my God, this thing has just figured this out for me, and now I can literally spend my time collaborating to try to imagine what we could do to use this to go beyond, right? So think in the supply chain example, but like think about now, imagine all the supply chain folks that we were just talking about, and now all the marketing folks. So I could put the fishing pole, I could put the right fishing pole on the right shelf at the right time and in the right quantity, right? But then imagine if I could actually work together collaboratively with the marketing and sales teams to make sure the person who would most benefit from that fishing pole. So I like to fish from a saltwater, you know, saltwater, but then the salt water from a boat. Then I want to fish saltwater from a boat with kids. Imagine if like I could actually anticipate what you need. The vast majority of business people are there really because they love their jobs and they and they want to do it better. So I think the um I think people's lives definitely change, but they it changes for the better. And um, and I I think that's a really exciting thing. And then, you know, the on the let's talk about the smart food thing again. We, you know, uh remember we're feeding the hungry, right? With it. So the you know, like the folks that are in the business of trying to feed the hungry, there is no more satisfying thing to see someone who needs to eat to give them healthy, fresh food that helps give them an opportunity, right? And you know, the over the last few weeks we've delivered over 27 tons just in in emergency bases into places because remember, the government was shut down and whatnot. And so trying to really make sure that people could eat, and you know, we fed 6,000 uh people, right? You know, thousands of families, um, all with food that would have been wasted. And I can tell you the universal reaction to that is oh my God, this is like, this is such a cool thing. So I there's the doomsday people with AI, and I guess if I had to pick a message, like you know, you want to get onto the train of, hey, we're gonna use this to make the world better, both our jobs better, make our companies more profitable. Um, and then on the citizen services and the government side, hey, we're gonna do, we're gonna stretch budgets and use our taxpayer dollars as efficiently as we can. And I I just think that that was such an amazingly emotional last few weeks as we're helping really solve the problem by finding surplus food. And uh that is a you know, the look on people's faces when they see the produce that they're getting that otherwise we were going to throw away.

SPEAKER_00:

Incredible, so important, especially this time of year. Um, uh when you talk about the enterprise, you you have a lot of legacy systems, we have a word for it, technical debt. How do you make an impact without breaking or ripping or replacing existing systems? It's kind of a jungle out there.

SPEAKER_01:

Yeah, so I think there's three secret sauces the way R4 deploys. A, the R4 can pull data together automatically. B, it's almost like settings on your iPhone, how we configure the AI technology now. So you don't need to understand any of the actual mumbo jumbo like your son is learning in his app building, right? But then the third part, just as the data can be pulled into to R4 automatically, the answers can be federated back. And I and I feel like this topic of technical debt, you know, we're never you're always going to need to upgrade enterprise systems, right? You know, for either, you know, new ERP systems or those kind of things. But what's so cool now is when we turn on the AI decisioning layer, think of it as like a layer on top that just allows all the stuff to be smart. And because we're not relying on that system to change what it's doing, we're just feeding it back an answer that makes it transformatively smart. It allows you to then get the ROI, which then you can then go reinvest in all the, you know, you know, the improvement and upgrade that systems may need. So I think AI changes this concept of technical debt because you it it doesn't, you don't have to rip out and build a data lake in order to get started. You can now with AI, leave everything alone, turn it on, make it smart, and coordinate it between the supply chain, between logistics, between the marketing silos and whatnot, all automatically, and then upgrade the systems with all the free cash flow that's produced. Because once you have the turn-on use cases, you know, it's uh you know how the IT works. You want to eke out improvement in economic um outcomes first, and that's what allows you to invest in systems.

SPEAKER_00:

I'll talk about a virtuous circle that's uh amazing. So you're already leaning into the future here, but what's next? Where do you see XDM and you know cross-enterprise AI heading over the next year or two? What are you excited about with your roadmap or you know, the industry in general?

SPEAKER_01:

Yeah, well, I'm we mentioned earlier, I'm really excited about V5 of R4 because it is just going to be even easier to use, right? And it'll be even more automated now. And because we've had so many deployments now across commercial defense and the civilian state, it's a lot of learning as to exactly how do people try to use the new technology, right? And then I think it's kind of neat watching, you know, we think what you know, how people will use it, but then to watch them actually use it, then now all of a sudden you can change up the user experience in a way that makes it even more intuitive. I think the um the complexity of the problems that we're solving has gone way up. And I think you're gonna see that as over the next couple of years, because in the old day, you know, even a couple of years ago, you know, you try to solve a complicated problem, but like this food thing, right? That is a really complicated problem. The data is all over the place, systems are crazy complicated. You you're now seeing us tackle these way more complicated, you know, problems, which involve more parties to coordinate, more left and right hands to coordinate. So I think that that's gonna be it. I think, you know, healthcare, I think is gonna be a big category right now, uh, that's new to us, but it's showing incredible promise, both on the public health side as well as the payer provider, you know, you know, the healthcare that we would think of. Um, I think that's a big area. I also think um energy transition and energy optimization is a huge area that you're gonna see coming forth in the next uh you know few quarters. Um and these are just you know even more complicated problems. I mean, it sounds like supply is complicated, but it's not nearly as complicated as you know the healthcare system is, right?

SPEAKER_00:

Oh, amazing mission, amazing vision. Congratulations on all the success, onwards and upwards.

SPEAKER_01:

Yeah, thank you. It's great to be here and great to be with you. And thank you for what you do to you know spread the word of new ideas, new thinking. And um, you know, it's it's time for everybody to jump in and um, you know, make the future happen, right?

SPEAKER_00:

Yeah, the time is now. And thanks so much, Paul, and everyone listening, watching, and sharing the episode. Appreciate you. And be sure to check out our TV show now on Fox Business and Bloomberg at techimpact.tv. Thanks, everyone. Thanks, Paul.

SPEAKER_01:

Have a good day.