Subscribe
Share
Share
Embed
Starting an AI company is all about spotting a real problem and using AI to solve it in a smarter, faster way than what’s out there today. It’s less about having the perfect idea and more about starting focused, learning fast, and building something people actually want.
This week, Dave, Esmee, and Rob are joined by Gijs van de Nieuwegiessen and Tijn van Daelen, founders of One Horizon AI, to explore what it really takes to start and build an AI‑native company
This week, Dave, Esmee, and Rob are joined by Gijs van de Nieuwegiessen and Tijn van Daelen, founders of One Horizon AI, to explore what it really takes to start and build an AI‑native company
TLDR
00:32 – Introduction
00:55 – Hang out: Why Dutch names can be a real tongue-twister
02:00 – Dig in: Exploring how an AI-native culture fits with human-to-human interaction
13:35 – Deep dive with Gijs van de Nieuwegiessen and Tijn van Daelen
1:01:54 – Following AI: Bloopers, reflections, and field hockey with the kids
00:55 – Hang out: Why Dutch names can be a real tongue-twister
02:00 – Dig in: Exploring how an AI-native culture fits with human-to-human interaction
13:35 – Deep dive with Gijs van de Nieuwegiessen and Tijn van Daelen
1:01:54 – Following AI: Bloopers, reflections, and field hockey with the kids
Guest
Gijs van de Nieuwegiessen: https://www.linkedin.com/in/nieuwegiessen/
Tijn van Daelen: https://www.linkedin.com/in/tijn-van-daelen-495986131/
Open source repo: https://github.com/onehorizonai/ink
Tijn van Daelen: https://www.linkedin.com/in/tijn-van-daelen-495986131/
Open source repo: https://github.com/onehorizonai/ink
Hosts
Dave Chapman: https://www.linkedin.com/in/chapmandr/
Esmee van de Giessen: https://www.linkedin.com/in/esmeevandegiessen/
Rob Kernahan: https://www.linkedin.com/in/rob-kernahan/
Production
Marcel van der Burg: https://www.linkedin.com/in/marcel-vd-burg/
Dave Chapman: https://www.linkedin.com/in/chapmandr/
Sound
Ben Corbett: https://www.linkedin.com/in/ben-corbett-3b6a11135/
Louis Corbett: https://www.linkedin.com/in/louis-corbett-087250264/
'Realities Remixed' is an original podcast from Capgemini
Creators and Guests
Host
Dave Chapman
Chief Cloud Evangelist with nearly 30 years of global experience in strategic development, transformation, program delivery, and operations, I bring a wealth of expertise to the world of cloud innovation. In addition to my professional expertise, I’m the creator and main host of the Cloud Realities podcast, where we explore the transformative power of cloud technology.
Host
Esmee van de Giessen
Principal Consultant Enterprise Transformation and Cloud Realities podcast host, bridges gaps to drive impactful change. With expertise in agile, value delivery, culture, and user adoption, she empowers teams and leaders to ensure technology enhances agility, resilience, and sustainable growth across ecosystems.
Host
Rob Kernahan
VP Chief Architect for Cloud and Cloud Realities podcast host, drives digital transformation by combining deep technical expertise with exceptional client engagement. Passionate about high-performance cultures, he leverages cloud and modern operating models to create low-friction, high-velocity environments that fuel business growth and empower people to thrive.
Producer
Marcel van der Burg
VP Global Marketing and producer of the Cloud Realities podcast, is a strategic marketing leader with 33+ years of experience. He drives global cloud marketing strategies, leveraging creativity, multi-channel expertise, and problem-solving to deliver impactful business growth in complex environments.
What is Realities Remixed?
From the award‑winning team behind Cloud Realities, Realities Remixed explores what happens when people, culture, industry, and technology collide, because real breakthroughs rarely come from technology alone.
In each episode, hosts Dave Chapman, Esmee van de Giessen, and Rob Kernahan dig beneath the surface of innovation to understand what’s truly shifting. They speak with leaders, practitioners, vendors and thinkers across industries to explore how technology reshapes human behaviour, organisational culture, and society, for better and sometimes for worse.
Web - https://www.capgemini.com/insights/research-library/cloud-realities-podcast/
Email - realitiesremixed@capgemini.com
(00:02.11) What's a doom David? This is going to be a happy fun experience. You're up the audience for the countdown of fun. That's better countdown of two. hell. The of the countdown to excellent and inspirational content. Yeah. Now you've over egged it now. We've never been inspirational and interesting. think you've gone a bit too far there. (00:34.158) I'm Dave Chapman, I'm Esmee van de Giessen, I'm Rob Karnahan. And this is Realities Remixed, an original podcast from Capgemini and This Week, a conversation about AI native organisation from an AI native startup and the reality of what productivity in the modern world might look like. Now joining us later in the show to talk about this and many other things are Gijs van de Nieuwegiessen and Tijn van Daelen.. And they're from onehorizon.ai and a shout out to Esz there for helping me with the pronunciation. Thank you very much Esz. Yeah, sure. You're okay with your Dutch, but maybe this one. They have quite some Dutch, very Dutch names. And sometimes you just have to know your limits, don't you? mean, the Dutch have certain words that you just go, how do you make that sound? I don't know how you do it. It's like this thing and you go, I don't know where that came from. It's a talent. Yes. And you did a fine job there. So if you want to If you want to jump to the conversation that we're having with Tijn and Gijs, you can get the... Esmee actually laughed at that. Yeah, I start to giggle every time you say their names. I'm sorry. That's cute. I am trying. Thank you. Everyone loves a tryer. Well, if you want to jump to our conversation with the guys from onehorizon.ai, you can find the time code in the show notes. Yeah, it's a good one. Big wide ranging conversation. So go have a listen to that. One of the things I wanted to pick up... as an interesting thread is this notion of AI native culture. Es, in your head, as somebody who spends a lot of time thinking about human to human interaction and the importance of it in the workplace, how does AI fit into that for you? I think it's part of a larger generational context. If we see the newest generations that are so digital natives, like the young ones that are, you know, they're touching iPads and screens, like it's their second nature. So AI for them is probably also going to be second nature and the generation we're, I'm a bit younger than you guys are. Let's not, let's not tell the numbers, but look, it's already been a rough show for me. It's already been a rough show for me so far, like, know, did you hear our producers turning 60 next year? And I was surprised. (02:55.822) I was surprised that... Oh, that's a compliment, I would say. Still young at heart, Well, you're surprised he managed to get to 60. I think that, yeah, think we're all surprised that he successfully navigated 59 and a bit years of life after we've had to have him deal with our logistics. But we're the generation that know, you know, we were the first ones using, you know, the Nokia, we're the first ones using telephone or Amazon or internet or et cetera. So we've got so much evolution. from not using tech at all to now having it integrained and even talking about it every week on the pod compared to the generations that don't know any better. So I think for them, it's also hard for us, or for me at least, to really grasp how it is for them to grow up in a world like this. And that's probably also gonna be with AI. Their generation is gonna make AI more, or give it the value that it hopefully can have. It's an interesting point, isn't it? raise it. It was the Walmart CEO basically said, I steered this company through cloud adoption and digital adoption. I don't have the energy or the will to do the AI adoption curve because he knew it was very different and he was strong enough and capable enough to go, that's for a different leader. My time here is, I found that quite profound and sort of like the willingness to say, no, this is a job for somebody else. Now I did this one, but I'm maybe don't have it in me to do that one. I really want to come back to that. in a sec, but before we move on from Ezra's point about the generational shift piece, one of the pods we did at Google Next, I think it might have been the one we did on creative AI. I might be misremembering that. The point that I thought that was really interesting was we are a generation of people who have learned to use computers to do our jobs. this act of using computers to do our jobs, it's that bit that might well go away. you know, the interface swap here no longer needs you to learn how to use a computer. And I thought that was a pretty insightful comment. When you then combine that as with the demographic point you're making there, there's quite a lot more increased fluidity in the interface, isn't there? Yeah, there is. But at the same time, we also hear that there are now countries having removed all IT now, even in the classrooms. They went back to writing, I think, in one of the Scandinavian countries. (05:22.86) So I think that's fascinating. it, we also going to turn back to an offline world again, just to make sure that we're, you know, that we stay in touch with the humans that we actually are. And is that maybe even going to make us feel better and happier? There's also got a scary point in there, which is not understanding the abstraction. So I don't know how it works. Proper sort of time machine stuff, which is... If I only know how to operate the computer because I can talk to it, you're basically creating a future horror if we forget how the machine works. It's just a thing that we have to trust. It's like, I'm not sure I want to go there. I would say that you are a Gen X who is technically interested and have made a career of being interested and understanding the mechanics in all different dimensions of computing. But I would say that the generation that starts using computers that doesn't understand 50, 60, 70 percent of what is going on to get that information to you is we're probably already neck deep in that, aren't we? And what happens if like there's ever a decreasing number of people who actually understand the machinery and then you go, well, if it breaks one day and maybe we've lost something, you go, is it Armageddon time? I don't know. It's a scary side to it. Or maybe we'll just learn to deal with it. maybe... drive a car. without really having the first knowledge of how to fix a car, if you see what I mean. Yeah, but you know how a cylinder works or a battery works or something like that. You know the basics, don't you? Why? Why do you want to know that? Because it's propelling you along at like 80 miles an hour or whatever and you're like, I appreciate the machinery that's able to do it. Or it gets me from nought to 60 in a very fast wrong person this question, haven't I? I've asked the wrong person. Yeah, but with the car, if you take that example, I really don't care what's happening underneath the hood. Look, we've got both perspectives, like right there, both perspectives right there. But before we lose the point, I wanted to go back to your point about leadership and Walmart, Robert, because I think there is, I think there is one, think it is a typically interesting comment from that chief exec who has driven a lot of change over. (07:40.032) over his career and you're right, opted out of this next lift. Now I wonder what's going on in that. So for me, when you think about AI nativeness, there are a number of different layers and I'll have a crack at a couple of them, but there'll be many more dimensions of this ultimately, I think, which is there's a layer that says, and we're gonna talk about this on today's show, how do you make the individual more productive? How does that work? How can you use AI to smooth workflow, to move things along, to automate certain sections of it? That's one aspect of the inevitability of this transformation. There's an aspect that says, once you introduce agents into that workflow, whether they're anything more than a, you know, kind of a repeating batch job, or whether they're more intelligent than that, how do you then treat that? Do you treat that as a tool? Is it more than a tool? who knows how that's going to evolve. So you've got this hybrid nature of how things are going to work and you know, of there might be knock-on effects of that societally and how does that work and what does that look like? That has to be brought into I think the leadership perspective and how you take that forward. And then at minimum you've got the HR and the finance organisations needing to then think about the financial model, how they're paying tax or the... the organisation and how they're thinking about their workforce, whether it be real or artificial, needs to be thought through. So even just that very basic sketch of the level of transformation that's going to go on, I can see what he means. That's going to be a lift, isn't it? Yeah. And I think I characterise it like this, which is if we think about cloud native changed how inside the box operates, the agentic AI future is going to change how the box changes the world around us and how work gets done. So it's a very, very different thing. box are you talking about? The compute box. We changed the way compute delivers digital and outcomes, the box, but now this is the box. This is the box. I've got the internet in the box, right? That's the box, right? We've changed how the box internally operates. What is coming is the box now affects the entire world around it. That is deep. (10:03.41) Come on. Inside. Does that mean we have to think outside the box, You went there. You went there. Very good. But that's it, right? Cloud native, change how the box operates. This is the box is changing the world around it. That's the way I think about it. Mic drop. I don't know what he means, but it sounded good, didn't it? Well, he is actually having trouble with inside the box with his internet connection today. maybe... It's very true. Actually, my inside box isn't working properly. You root the box. to the internet box. Yeah, the technology has failed. It's not failing. It's just going like slow like my brain is on a Friday. am looking forward to the weekend. I must admit it. What are you excited about doing at the weekend, Rob? Nothing, literally. Christine's going on a yoga retreat. So she's off. No, we all know how much she yoga. Without me. That's like, hey, I would never do that. That's like. I know. So I think I'm going to just relax. Yoga retreat? Yeah. Absolutely. A friend does yoga, Gemma, and she goes along and they do a big managed day thing. That sounds amazing. Well, it might sound amazing for you. Christine loves it. So she's off. Brilliant. And then I'll have to work out what to do and I'll probably end up cooking steaks or something. Who knows? It's like a natural result. We brought back our... When we were in Vegas a couple of weeks ago, Marcel, we went to a barbecue place, as you might... have heard on the shows that we made there. Roadkill Grill. Roadkill Grill. And Marcel and I brought back some seasoning from there and I was looking forward to it. So the weekend I used the seasoning and I made it, but I didn't read that it already had salt in it. You double salted. It was like, we were all being polite, but it was the saltiest thing you've ever put in your mouth. Don't Rob, don't. Rob's actually put his hand over his mouth. Very wise. Anyway, next time I will admit the additional salt, but it was delicious though. I will say that it was delicious. You just needed to drink three litres of water. As well, we're on it. What are you up to at the weekend? I've got to mow the grass. No, it's because of the great weather and the garden is fully exploding, which is great, but it needs... (12:26.36) quite some work and the weather looks good for this weekend. So I'm really looking forward to do that. Spend a lot of time in the garden. How about you, don't forget yourself. Yeah. I'm also having a quiet one this weekend. We're going to see some friends on Saturday night for a BBQ. I might take my rub with me. know, have rub, we'll travel. I love the idea of you turning up to somebody else's bar and you go, I know you're cooking the food, but I'm going to interfere and change all the flavours on you. I mean, that's classic Dave, that is. That is the, that's the kind of value I like to bring, I know you've got everything set, but let me alter your world. Exactly.
Well, look, on that note, enough of all of that. Let's jump to our conversation with the guys from OneHorizon.ai on AI Native Culture. and many other things in a really fascinating wide-ranging conversation from some people who are just starting a journey in an AI world. (13:35.97) So Gijs and Tijn, thank you very much for joining us this afternoon. How are you guys today? You're welcome. Doing all well. What about you? Yeah, doing great. Thanks. Great. Great. We're okay. Whereabouts are you in the world? Well, we're both located in the Netherlands. More to the south. So I'm close by Tilburg and tying this new wildlife. We're massively outnumbered by the Dutch today, Dave. There's now four of them on the call. They're all like, ganging up on us. is... We're a minority. We're a minority in many ways. Tyne, what about to you in the Netherlands? I'm in like it's like closer to Tilburg as well. If you don't know the Netherlands, like it's in the southern part, right? It's some people they know from the world, know Eindhoven. It's about like 20 kilometres from Eindhoven. close by. Right. Very nice part of the world. welcome and really interested initially. Why don't you just... Tell us a little bit about yourselves and your organization. Yeah. My name is Gijs. We're One Horizon. So a startup previously before starting One Horizon, I had another startup, which was Float.ai and we sold that in somewhere in 2020. That's how I also met Tijn. He joined that company. We got acquired by Vista Equity, a US based investment firm, and we joined a US tech company. and worked there a couple of years. But yeah, it started to itch again, to start something new. So that's a bit how One Horizon was born, passing it to you, Tijs. Yeah, like Gijs mentioned, I started working there initially for marketing. That company I started was doing conversational design, so basically building chatbots for larger brands. And I liked doing that a lot more, actually. So eventually grew to that role, started doing professional services, and then later there, became product manager. Then once It was solved, like Gijs and I, like Gijs mentioned, started getting those issues again, and we really wanted to solve our own problem. So that's how we got into One Horizon. And what we basically do is we make the software development lifecycle easier. Like everyone knows that cursor and clouds are changing the way you write code. But why only change that part, right? Well, it doesn't really make sense anymore to do everything around it in the traditional way from before Gen.AI. (15:57.602) So now we help with the planning, the communication, reporting to your manager, sharing release notes, trying to optimize that entire process. Well, we'll come shortly to the specifics of what the organization does. But what I thought might be worth talking about a little bit to start with is the sort of the process that you guys have been to to get from the kind of initial idea following the sale of the last organization to a point where the company is now functioning and running. So take us back to that post-sale moment where the two of you were sitting, I don't know, whether you were sitting at desks or just thinking about this problem. What was that initial inception conversation like that you decided that you had an idea you wanted to act on? Well, we worked a lot with international teams when we worked for the US-based company and we saw a lot of inefficiency there. we were like even When we got bought, were a small organization, but we were very efficient in the way that we operated. And when we joined the larger organization in the US, we thought like, hey, like these guys, big tech company, right? They work for amazing brands. They're probably a lot better at the stuff that they do than us, but it wasn't really the case. Like they were very inefficient. You're not suggesting that large corporates don't do everything perfectly, you? I mean, that's just insane. It wasn't even that large. It's a company, right? So it's like 200 people in product. So it's not a super large organization, but still very enterprisey and a lot of process and red tape around it. well, Tyne and myself, we talked about, how can we optimize the way that we build software? And along the way, you had that new revolution, right? Where you had the first launch of GPT that you could actually start to use. And we saw like a big change coming in the way that you build products and the way that you think about AI. like we have a very long background in AI and making AI work in a controlled way. So we were like, hey, we see like an opportunity to fix this for ourselves and productize it. (18:22.754) so we can also help others with it. That point you make on an organization of, you know, maybe less than 20 people, then to use your phrase, as it gets to 200, it starts to processize itself for want of a better term and create systems of work that help run the organization. then to Rob's exclamation of up to enterprise. you know, how do organizations self sustain? What does it feel like for you on that journey? At what point do you think an organization tips from the agility and purpose of those initial founded organizations into something that then needs to create systems to sustain itself? And I think I was hearing in your point, but correct me if I'm wrong, that maybe there's something gets lost at that point. Yeah, I'm not sure if you really have like a turning point somewhere. Typically what I always said was once you hire lawyers, like you hire a lawyer, like internally, then you've become that kind of company. Right. What a great KPI. Once the lawyer turns up, you're like, right, we're nobled. That's a bellwether, isn't it? Yeah. So, and of course, like in the U S everything is done with lawyers, right? So it's, slightly, slightly different there maybe, but even for a company that we work for, like 200 people in a product and they had like a whole lot of people in services and other divisions, still they, yeah, it was very slow in deciding and what to build next and to roll out new features. And it also had to do that once you start merging companies, you get all that, that extra baggage and yeah, that, that history that you need to maintain. then if you work for larger companies, like you have Microsoft or whatever as a client. Like they all request and demands that you, you have certain as a lace and, and, and, and specific customer agreements. And that all just piles up on it. And I think also what you see today in light of AI developments is you see a lot of people being able to produce a lot of greenfield stuff like, Hey, let's create some new, new app, whatever. You can do it really fast, but for these kinds of organizations where you have like a brownfield. (20:40.91) situation where you have like a lot of legacy code and whatnot, it's a lot harder to innovate and move. What do you think as you get to those points that maybe there's some of those things you set out, I maybe practical realities of doing business at a bigger scale and those sorts of things, but what would you, if you were scaling an organization to that level and perhaps did in the past, what do you seek to culturally retain through that process? What do you hang on to? in terms of values and ways of working that you wouldn't sacrifice because you have to do certain things. Yeah, that's really hard because if you start to grow, you have to hire other people, right? So you get extra layers. Even when we joined a large company, we were able to keep our culture intact. a lot of people, like we were, we had a culture of you work hard, you play hard and we learned from each other. And we shared every pain. across the organization. So whether it was paying within developments or it was like you lost some sale or you want to sale, everything was shared within that group. And that really helped in all employees, even contractors to feel and have like a sense of that they were part of that company and they were owner in it. I can imagine like you're talking about inefficiency and at the same time you're talking about sharing a lot. And I hear a lot of companies that say, they keep sending emails and they're overwhelmed with all the communications. Just let me do my job. that's like a, that's a bit of a paradox, is it not? If you say we're automating or remaking life easier for engineers, I would think that's what One Horizon helps doing. And on the other hand, you say like, but you want to share. and feel the gain and the pain of everything that we all experience throughout the day. Yeah, it depends a little bit on the way you do it. Like if you demand a lot of input from, for example, your developers, then you're making life worse in some aspects. But if you're more proactive to them and share more updates to them, like positive news, then it can be a benefit. for example, we have a Slack channel called Winners. Whenever a new person signs up, everyone gets that message. (22:59.126) after some time it goes more active, right? You get more and more signups. They don't need to do anything for that. There's developers, but it's still giving them motivation. Like we're on the right path. We're seeing more and more signups. So these kinds of insights, you want to share these little victories together. So that could be an example of something that doesn't add work, but does add like wins and more communication in a proper way. I'm also interested before we move on to One Horizon particularly, I was interested in your take on AI native culture more generally. So if I was to frame it up, you know, we've gone through a period of, I'm going to call it traditional tech or IT cultures framed in the world of client server and the lead times that that took and the ways of working that that took. And then we went to the world of digital or cloud native and that had, you know, very specific different ways of thinking. It was connected to agile and iterative work and moving quickly and experimentation and learnings from failure rather than longer period execution and waterfall execution of projects and programs. I wonder how you would describe AI native culture and is it profoundly different to what we might have called digital native previously? Is there something different going on there for you or are we still in the same cultural paradigm? If I talk to two of us, like we see a large shift there and we also see a lot of companies struggling with what it means. So there's a lot of CTOs and engineering leadership that gets pushed, right? From the boards, from the CEO, like, look, we need to adopt AI. How are we going to do it? And they really have no idea. Because it's really like it's a change management process and it took us years to even adopt agile, right? Like you can even argue that a lot of companies never adopted agile. They just work waterfall because that's natural and the same will happen with the AI. Like you'll have new ways and processes that you need to implement and you'll come up with new roles within your organization and how you're going to apply it and also how you're going to track in what's happening. (25:21.912) Right. Because let's say in five years, if you believe some people will have AI agents take over responsibility and doing stuff fully automated. That's certainly the hype, isn't it? The hype is that within, you know, some large organizations would claim you could do even do this now, but let's even say three to five years, you've got these human agent hybrid organizations that are suddenly you know, kind of massively automated from where we are now, those sorts of things. Is that the sort of hype that you're referring to there, Kees? Yeah. Yeah. And it's, it's, it's sort of like when you, when we introduced computers, right? Like we started to work more efficient, but look at the number of people that we have working in doing paperwork, right? Like you have automation, but it, and especially now and when we talk about it. You get AI and you have suddenly you have like magic in your hands where you can be even more efficient yourself. So instead of taking it easy and doing the same amount of work in less time, you're actually doing a lot more work because you can do a lot more work. And that's other side of the story, right? But for organizations, like how do you determine who the right skillset is for that new future? That's extremely difficult. It's changing so rapidly. I don't believe in that hype. You just described like, hey, let's completely replace departments. But that being said. What we like two years ago we worked, we had people working just purely on documentation right within our teams. They were writing up documentation whenever a new feature was ready. Right now we just have an AI agent that is taking any new changes. Creating like a request to update documentation. We tweak it, we approve it and it's done and it's way more efficient than we used to work. Right? And we no longer have to hire that role. So yes, you're definitely going to see changes there, but what you do need is you need someone that has like a very good understanding of how to set up documentation. How do do information architecture and make it really good for people. (27:45.356) right to understand because they know a lot more around the context of sharing that information. So it's a different kind of role and a different kind of skill set as well. So you're definitely going to shave off a lot of human hours there, I think, but yeah, there'll be different roles that will be created within the companies. when you talk about documentation, it's a good example of where two years ago, if we tried to implement it, it would have been really Janky, incorrect, know, have loads of issues. Now the efficacy of the AI implementations is a lot better where it is literally maybe a couple of tweaks on the way through, but broadly this is it and it's done and it's taking huge amounts of activity off the plate of humans and there are activities that humans were crap at anyway. So it's the classic agile was always interaction over documentation, but it wasn't like... don't throw the documentation away. You still have to do it though, don't you? Do ever bump into one of those technical writers? know, we said technical authors, Yeah, yeah. was a job. ever bump into one of them that are clearly a frustrated actual writer? Yeah. of fancy themselves a little bit as being like, I should be writing novels. Like I should be writing the great American novel, but actually I'm here writing this tech manual about Assembler. It's the point though, isn't it? We used to have a job to write technical documentation because we knew it was hard to do. Now the AIs are doing it for us. That job's redundant. It's like that joke, which was I became a prompt engineer, then I did something else and then I've ended up as a farmer because the AI went so fast. A job that was now thought of being as the future is now it lasted 18 months and it's gone again type thing. you're right. And so what do you think, maybe Heiss, what do you think organisations who are shifting from, know, or trying to get their head around what AI native culture is. Have you got any thoughts on like leadership style and whether leadership style in those organisations needs to change? Or do you think that what we're looking at here is like a process level change in the main? Well, you'll definitely have leadership change. But like, the weird thing is, the way that I looked at it is (30:02.518) Almost the first role you can automate away is a CEO. And this sounds really weird, right? But that person is like taking responsibility, making choices and communicating to the boards and taking responsibility to the boards. But that's still a person, right? We keep it there. But why is that a person? Why is that not AI? And that's like also a thing, like a way that you can think about what AI will do, because I'm not saying you don't want to have person there being responsible, it's more that when you look at making decisions when in a business, and if you're really AI first, why not pull in the data then as well and take away that human politics and everything around it that you have in large companies, right? Just think about that kind of scenario. How does that land with you as an organization with sort of human politics? maybe not fully stripped out of it, but stripped back a bit. that something that you think, that sounds attractive and cleaner or is there something missing there? Well, in essence, think we're all here on the world to connect with others. and politics, you know, there's power everywhere. So I don't think that's a bad, and that doesn't have to be a bad thing. Because what Gijs, I think, is mentioning is also about transparency. then at least if the data and AI is working for you, that can be very helpful. Also to communicate throughout the entire company, you know, one dashboard for all, not a secret dashboard that only the CIO has or C-level has. So think that's good. But I think it opens up the possibility to have people in leadership roles that are truly leaders and not managers. Because I think what you see now is that a lot of leaders are actually managers because they make the decisions and they're very control-based. Very figure-based making decisions. But if you take that away and you can fully trust AI with that, then you can have true leadership that is a person that connects and that networks and that, you know, opens up possibilities for potential talents and really see them. So that could be, I think for me, that would be like. Fascinating take. I don't know if it's a better world, but at least, yeah, maybe it is a better world. The automation of middle management. (32:23.95) That's a fascinating take. Well, just rejecting everything that comes into your queue for authorisation, you mean? That's pretty easy. just tell a bot to hit no. it at a gigantic speed, not human speed. Yeah, Machine speed rejection of everything. I think that's, I think we might be able to cope with that these days. Do think it will just iterate the nose so fast, get so pointy way it just shuts the company down? I love the idea of there's a spinning wheel and the whole company just collapses in about two. Exactly. was inevitable decline of this organisation. switched on the automation of the middle management system and it shut down the company in five minutes. Yeah, rather than five years, we took a five-year collapse down to five minutes. That's it. I mean, that's a good KPI, isn't it? Massive performance improvement. Well, look, think we've successfully got to the end of that thread. On that note, then, let's move on to One Horizon specifically. So Tijn, why don't we return to that moment of the inception of the conversation of you guys wanting to go into business together again? What problem were you trying to solve? Yeah, so we were getting pretty frustrated actually by those inefficiencies. Like we worked in automation, it's basically our goal to make everything as automated or as streamlined as possible to scale it better. And what I was doing as a product manager was just passing along information like half my day. So asking that developer, like, are you already done with that? Telling another developer, like, hey, here's this update. You can continue with this part. Then getting back to other stakeholders. And that's not really that productive, So you're feeling sometimes a bit productive because you put something in a slide deck or in some Excel. But are you really adding value? I don't really think so. So just generally on that point before you go on that I don't think there's, think if you, if you have the mindset that says, you know, I'm going to, I'm going to track every keystroke that I do from a point of view of saying, is it like a value add keystroke? Like, am I doing it for a client or am I doing it for my product? And then you count every keystroke that's about filling in a time sheet or (34:40.534) writing an email to somebody about something about that other thing we had to do about that thing. Like it gets depressing really quickly, I think initially, but then if you start to tilt your time spent in the right direction, it is to the good. So pick up the story there. I just wanted to illustrate that point because I think it's such an important one. definitely. Yeah. And then if you don't change that direction, people will just... spend their time on writing some scripts that presses more keys, right? So they look even more productive, they get even less productive. So I was pretty annoyed by that inefficiency, starting with communication and now, of course, we were in chatbot design, so you already experience with designing for conversations. But then once we got so many more opportunities with Gen.AI, with all these advancements, we were like, yeah, we can build some dashboards, some platform. that takes away a lot of that unnecessary communication and puts all that information in a single spot. So that's basically how the initial idea began. So just give everyone the single source of truth. So you don't need people like I was before to pass that information along. And how does that look? So I'm a developer. God help us if you are. I was just role playing. It was too easy. Yeah, okay. And I was just trying to think, how do I describe me being a developer? I think when you're all play, it's supposed to be believable. Small town. I'll tell you what then, let's do it different way. Rob's a developer. Oh God. Now Rob's one of those developers that constantly bangs on about how good he is about stuff. He's annoyed about the fact he doesn't work for NASA. I'm ace. Yeah, yeah, yeah. It still goes on about the halcyon days of writing Assembler and he needs a tool to make him more productive. So Tyne, from Rob's perspective, how do you help him with this? What does his experience look like? To basically let him stay in flow state longer. God help us Tyne. That sounds dangerous. Can you imagine if I was in flow state permanently? I can't. I can't. Yeah, like when you get distracted, it can take like more than 20 minutes to get back in that flow state. (36:58.554) And you're already distracted by things like your phone all the time, right? You can put it away, but then you still get distractions from work, like from emails, from Slack, or a lot of meetings you have to jump into. So if you can just automate like all these small notifications, or just give the person that asks for it, the information without having to ask it, then the developer can just do that work. And the manager can just ask like a Slack bot or... some other dashboard they can use to see what that developer has done or what the update is. And that takes away a lot of friction there already for developers. They don't get distracted that much. But also it stays more truthfully. Like if you ask me something and I ask guys for it, then everyone changes the outcome a little bit, right? And at the end, if you pass it along enough, you get a completely different story. So if you just have all the information you need in a single spot, everyone could just look at it whenever they need without distracting anyone. You already have a lot of gains. So that's the initial thing we started working on. And then once we had that, we kept building upon that. We've talked about this before, Is it now, are we getting closer to reality in the data? And is that a good thing? Because we also talked about the politics and their reasons why stories change. And that also has a cultural value. to it. What do you think, Thijn? Is it better if we get closer to reality in data and in the things we share? Much better, I think. Because if you remove a lot of politics, you also speed up the processes a lot. What we noticed at a former company we worked at is that the politics especially slowed down everything. You had to plan meetings with executives or other managers. They were very busy, so you could only schedule a meeting once a week. then some minor decisions sometimes took like a month even. Now with AI, everyone's moving fast. So if you keep these politics, you're always too slow. But there is a part in that though that's quite important is this human nature when you build sister. So when you have a small team, human nature is different. Going back to that point about the efficiency of a small team, joint purpose, we all feel the pain, we all feel the joy. (39:13.698) And then when you go to this certain bigger team, inherently human nature kicks in and politics arrives. So I really like the idea of trying to remove it and simplify the journey and get that. But there are certain things in life you just cannot escape when you get to a certain amount of people and a certain communication structure. It's just the way we are, you know, we're wired into that. And I think it's noble to attempt to remove it because when you do, it does simplify things so much. Yeah. Yeah. And the goal isn't to get it perfect. That's, think, impossible. Like you're right. There will always be a bit of politics, but that's okay. As long as you just improve it drastically, that's already a massive gain. It's always like how you share something, right? But at least you're going to share something based off facts. That's, and it's your decision that you got to change the facts a little. That's fine. But it's like the thing that we noticed is in any company, when you build software, you basically have a plan, right? build and then in the end you share or you report on it. And if you look at larger tech companies, you sit together somewhere at beginning of the fiscal year, you're going to share your roadmap. But yeah, we're going to build on this, share it to the board and they're going to agree and not agree, whatever. And then you're going to go back to your product teams and you're going to articulate, guys, this is the plan. This is how we want to grow revenue or whatever OKRs, KPIs you have. and you figure out a road map, then that road map is pure fiction. It's never going to happen because you have people in your engineering department. They're going to say like, it's great that you're going to build an enlarged other app, but we still have this tech debt to solve. We're never going to launch a new app if we don't do this. And that's not being articulated to the board, right? So what with one horizon or solution like us, what you're going to get is you're going to draw your roadmap, but you're actually going to roll up any work that happens within your company is going to roll up back to that roadmap. If you're going to share it slightly different to your board, that's up to you, right? But it's going to give you at least, it's going to give you like actual what is happening within my teams. What are we actually working on instead of like just making stuff up? (41:39.21) It's always felt to me, I love your example there, the way that nests underneath roadmap and then how you share roadmap. Because it seems to me that every time you queest something slightly differently in the communication, just creating more layers that have to be then translated going forward. So where you might have had a very simple roadmap with milestones one, two, and three on it, and we're going to deliver them. first, second and third quarter, whatever that looks like. The second you've nuanced that or changed that or something to create a nicer narrative at the board level, you're setting yourself up for a problem at that point, aren't you? Yeah. And it's also the translation. And that's the other thing that AI is great at, right? Like AI can explain a very complicated piece of code to my mother. Like that's what it can do. So one of the things that you can also do is you can... take something that you as product or business aside on, like, we want to build an app, but it actually, it breaks out in all these very complicated technical components. But people in products, they don't understand the technical components. And the people in tech, they're like laser focused on that very specific problem that they need to solve. And you have something in between that's the glue. And that's like one of the benefits there. So one of the things we were kind of alluding to there is context. And I wonder in your platform for a developer at any particular stage, how you're managing to establish and track context. Yeah, any work that you do just rolls up in that context. And that's also the other thing that we see, and especially in a couple of years, is you will have no idea as a human. to actually understand the entire complexity of everything that's happening. Why is it happening? Who did it? And why did it happen? You basically focus on the actual business problem that you try to solve and not all the smaller pieces behind it. And it's already a problem today, right? Because if you have a tech company that exists for 10 years, you have very old components of people that have long left your company and no one knows why it was made. (44:01.664) It just has like a comments above it in a code, is do not touch. That's it. Right. Now, you, probably, I'm sure you'd never find anything like that now in, in human derived code with you. Well, probably AI is even like adding like these kinds of comments. That's probably because it's all based on humans. Humans don't Yeah. You often see also that the AI just removes all of comments, even if it's useful information. And then if you add a comment like, do not touch this and the AI removes it, can also get into tricky situations. You can get into tricky situations. Maybe to bring our conversation to a bit of a head for the day, I wanted to talk about a tricky situation, which is the nature of competition in the world within which you're operating, where applications and platforms can be popped up. Heiss, what's your take on operating in a world like that? Yeah, it's like it's so different than when I started my first startup, right? Like the to your point, it's it's really easy to just generate something and throw it online and share like, look, we have like solution A, B or C that does this. But in the end, like writing code is just a very small part of building a business. It has so much more there and. The other thing that you also see and it's also a trend that you see, like I saw the other day on LinkedIn, there was this guy, said, like, look, I have a bill of $5,000 a year, right, for Slack. I'm gonna vibe code Slack and I'm gonna do it for a hundred bucks, right? Now, like, and I'm looking at it, I'm like, okay, well, an Apple developer account is already 150 a year if you wanna build a Mac OS app, so good luck with that. But it's like you're going to build that entire thing, right? Then you got to support it. And even if you have like AI agents like what the hell, it's not your core business. It's even like we're in this app right now recording ourselves. Why should we five code it like it's a couple of bucks. Someone's maintaining it for you. Someone is doing all the security around it. (46:18.486) Et cetera, et It's so, there's so much more than just writing some code, having like a nice vibe coded app that you can, you can share on LinkedIn compared to building a business. That being said, it is a lot harder now if you look at fundraising and that kind of stuff, especially in Europe. US is a different story. With vibe coding. people who got into it and they found it easy, they haven't always understood the nature of software. So software is not a static thing. And like you say, it needs to be continually addressed, updated, managed, supported, you know, just from like zero days and things like this. And people... this perception that I'm going to vibe code something, stand back and go ta-da, that's it, product on shelf, I'll move on to the next million dollar idea. it's like the world of software is somewhat more complicated than the talk to an AI and create something. And it's that stark realization that goes, nice idea, however, dot, dot, reality bites. So yes, AI can help us a lot, but there's still a practical nature to software that we must serve. is you, Rob, think, speaking from experience after your vibe coded Tetris didn't catch on, innit? That is my AI test now. getting pretty good, but each week you need to update the Tetris game because the code, the AI gets better, the code gets better, the game gets better, you update it. People want more features all the time, isn't it? If it doesn't change, people get bored of software. That's a big lesson there from Rob. Quick question. Is Block Blast better than Tetris? No, Tetris is a classic. Tetris is the OG, think. Yeah, of course it is. It's just that. I had Tetris on the Game Boy. when it originally came out and it was like, blew me away. It was just such a fond memory of a simple game that can steal hours from you. Block Blaster's a bit old-flashy and they've corrupted the idea. Corrupted the purity of Tetris. Corrupted the purity of Tetris. I'm sorry, but some things are perfect and should be left that after me saying about the software constantly has to update, I'm saying that bit should be frozen in time. That's also a cultural thing because the moment you say Nintendo... (48:22.586) I just, I see myself sitting on the couch again using that Nintendo playing Tetris. So there's a whole, you know, experience with it. takes you back in time. It's the old retro gaming craze that keeps coming back and uplifting. Everyone talks about Snake as once you say Nokia, we say 313 in Dutch. Just a phone that you use as a brick. Well, the battery used to last for two weeks though, wasn't it? There was And you can drop it in the toilet. You can just drop it on the floor. Really great features. Yeah. Actually a friend of mine, he dropped it from the roller coaster ride in the Efteling, the Python, and it dropped on the ground from all that height and it was completely fine. I mean that, that's a test that all mobile phones should be put through, I think. Fling it off a roller coaster. mean, where Ez went was I was just going to say just drop it, but you added the toilet part to it as well. I don't know. you know how many women got their phone dropped in the toilet? even in my small circle of girlfriends, I think we're up to 80%. 80 %? It's not mentioning names, but yes, just a note to self not to ask to borrow the phone of any of his or her friends for a phone call. I think that's on the list of things to remember not to do. I'm not sure we end how we ended up there, let's get back to Tetris. It all happened. Tetris. Yeah. Just before we bring things to a little bit of a close, let's just talk about challenges. It is a complex, fast moving world. We're surrounded by speed of innovation. We're surrounded by noise, communications noise from social media in particular and various different things. So want to just unpack that a little bit. Let's maybe talk about the nature of noise in the market time to start with. So how are you actually communicating what you're doing being the beginning of a big journey, needing to get it out there and stand out? How's that feeling? (50:33.998) Yeah, trying to add a little bit of a personal twist. Like we see it's not really that efficient if you only do corporate style marketing, right? So especially on LinkedIn, we also share a lot of like pounder insights posts and all. But what we do see a lot is that everyone is just flooding those, those medias. Like you get so many vibe coded products that people are trying to advertise on, like comment agents to get access to my agent, right? And everyone is spamming those replies. So you don't really get those useful conversations anymore. Also with blogs, it's not much easier to generate content that's not that clearly AI anymore. There are much better ways to do so. So you see like on all these different channels that also on Reddit, for example, that people are just automating it. And because everyone's doing it, you have to find a way to do it yourself in a good way that's not too AI like, otherwise you can't stand out anymore. So yes, you have to be original and authentic. So just filming yourself is already much better than creating some nicely edited video, I think, these days. But also you need to mass produce in a fast way because everyone is moving so fast. you do, while having to be authentic, also need to find a way to create a lot of content. Are you finding that you're generating the engagement that you want off that style? Yes, we're growing, especially on the SEO side, I think it's going well. That's mainly because of the blog, also some link building with it. But something we've been working at is like some own agent. We've made it open source as well. It's called Ink. And what it does is it's combining a lot of different skills for an agent to optimize all the different steps in creating a blog. And then it does what we call a Gordon Ramsay review. So it roasts everything about it, because AI is always way too positive. So if it's awful, it says, yeah, this is perfect. It actually manages to get the quality to a much better level like that. And nowadays it's already better than what we can create manually. So we just generate a post on a daily base with some automation and codecs. And that helps a lot actually. So even if it's Gen.EI, it's still working. good. We have some similar challenges, don't we? As in terms of like when we, you know, we we've had a number of different shots at trying to create social media. (52:59.374) just around the podcast, for example, and it's a challenging world out there, isn't it? Yeah, it is. I think we all started smiling because we're trying to do the same, just to make personal videos, one-takers, and we'll see what happens and make sure that it's not too ridiculous that we're showing ourselves like 100%, especially some of us. looking at the... Oh, I you were responding. was looking at you, but we're doing this digitally. it's even... I was deliberately trying not to respond to that. could feel it coming my way. was like, Oh, that's the human side. yeah, no appreciation. Well, and I think this is also and it's also what we hear as feedback on the pod and the same on the social media. If it's authentic and people feel that, you know, we're just being us and you're part of the hopefully you feel part of the conversation that resonates. that stands out as we see in the feedback as well. Humans like listening to humans, don't they? I think there's a real difference and I think it shows up in your time, the sort of videos you were talking about and maybe the tone of voice in blogs and certainly what we try to do on here is you write as it's humans for better or worse. And I think that's really different from say consuming notebook LLM where, look, you're going to get a really good aggregation of information on something like that, but you're not hearing the points of view or the fallibility's or the attempts at jokes that humans would try and do in these sort of situations. Yeah. We try to implement that into that ink solution is it has a lot of context about you on your personal level, about your businesses you work for, like your work experience, which are linked in as a starting point, for example, but also a different context file about your hobbies, your past experiences. people that inspire you and then it uses that context to personalise the output. That was maybe with Gen.ai, but it's still very true to yourself and that improves the content as well. it's the crossing the uncanny valley, isn't it? At some point we'll start to get much closer to humans or the way humans behave and then we'll have an interesting debate in society about when it becomes unperceptible if it was agent or human, where do we go next? (55:16.558) Because at the moment you can still tell broadly, Kanye, it might take a minute or two, but it becomes obvious at some point. And then someday it might not. Indeed. Indeed. Heiss, something else I think that came up in one of the conversations we were having when we were talking about making this together is the notion of operating in a globalised world where tech innovation in different sections of the world are moving at different paces. Give us a perspective on that and what you think, what challenges does that bring? Yeah. So for us at least, like there's, there's two major challenges. One is, um, fundraising, which is extremely difficult in Europe because the way at least to how it works in the Netherlands is everyone is pushing back like, okay, AI is happening. No one knows what the future will be like. So we're not going to invest in any software company. And in the U S what you see is they just. do buckshot type of investments. Let's just invest in like 20 companies and one of them will grow. That is like a challenge for us, right, in raising money. The other one is where is control? So our solution is also based on LLMs being built in the US. You have some in control by China. And I do believe that along the way, it might take like five years. you'll be able to run some of these models that are good enough for our solutions, just on some hardware. But still, like, who do you trust there? And it's extremely costly to build up your own model, right? It's almost undoable, especially for a startup like us. But even as you look at Europe in the situation that we are, and we want to innovate altogether, yeah, we need to act like really fast. in that regards, yeah, everyone will, everyone like us, like a company like us needs to move to the US in order to survive. Which is, which is a major, a major sovereignty problem amongst other things, isn't it? Yeah, definitely. And also if you look at a company, your company is the people, right? Like that's, that's it. And those people, they have like all the context around your company and all the knowledge and a lot of that knowledge will go now into agents. (57:41.282) So where is that knowledge going to live? Is it going to live in mainland US or is it going to live in Europe? I think if you look at all the politics that's happening right now, it's also for larger IT companies, but even banks, insurance companies and everyone, like something to really think about. when you, when you look forward, Antoine, when you look forward to maybe the next 18 months of development. and product roadmaps that you've got. Where would you like the product to be? You mean our own products or like in general where products are grown to? I was thinking particularly your product, but if you would like to bring in some of the wider sort of context to it, please feel free. Yeah, I think there's some overlap as well. Like we already discussed a little bit that you have that feature creep that people just keep adding more and more features. We think they'll expand to something we talk about as product creep where you see companies building new products even instead of just features. Think of Notion as an example where they built a full email solution or a calendar. you often do that. One of my pet hates, by the way, one of my absolute pet hates is products that tag stuff on and then have exceptionally inelegant interfaces as a result of that and don't integrate properly outside of it arc. Do you mean like Apple? mean, I think you've just described Apple. Don't integrate and just bolt loads of stuff on all the time. Absolutely the exact opposite to that. Sorry T, keep going. no, it's fine. I think we'll see that a lot more and also done mainly with agents, right? Because you get so more productive or you feel so productive with them. And therefore I think a big shift will become to properly take control of these agents and really determine what do we actually need to build so it doesn't get out of the end that we don't become the next Apple in that bad way. Of course, in some regards, I would like us to be the next Apple. I think that is broadly preferable on a number of different levels. Delusional. But like some direction I want to grow to with One Horizon is to be able to orchestrate these agents, control them, see their output. So make it a little bit more mature future-proof in that regard and really set up like all these new vibe coding startups or even like larger enterprises. (01:00:03.074) to properly use AI and agents. And Hai, do you have anything to add to that in terms of where you'd like to see the business going in the next few years and also perhaps what you're anticipating seeing in the cultural development of your customer organizations that you might want to play into? I totally agree with Tai. Now for organizations, like in a couple of years, we definitely believe, right, that there will be a lot of agentic stuff going on and across the board. So for us as growth, we would maybe even go into other departments like procurements or marketing or like departments that are close by what we're currently doing with basically development. What we try to do with One Horizon is we don't try to create another dashboard or anything, just create like this invisible layer that you connect and you keep up and running. And then you have all your reporting in place for larger organizations. Now, what we're currently focusing on is really on tech companies. Most of them are in the US, right? That's just a fact. But we also have like a very strong position in Europe. And we also believe that in Europe, there's going to be some shifts. At least we already see some shifts happening where the traditional players like Microsoft and others are being thrown at like, okay, Do we really want to go with a US based company, put our data there or not? So for us also focusing on Europe is also an important piece of our puzzle for in future. (01:01:54.7) Now we end every episode of this podcast by asking our guests what they're excited about doing next. And that could be you've got a really fun restaurant booked at the weekend that you've been looking forward to or something in your professional lives. So Tyne, what are you excited about doing next? Yeah, I'm just always following these advancements in AI. And it's just so inspiring to see like every day there's something new. People are getting so creative also with some ridiculous things with open claw and all. I just really like following that, seeing that creativity. just curious what will happen next. do you enjoy following the AI bloopers with things like OpenClaw where people ship it to production for some bizarre reason and then it destroys something? Yeah, there's some website for that. What's the name again guys? like... Agenthorrorstories.com (01:02:48.91) Yeah, I saw that one. Yeah. And the backups. That was the bit. It wasn't just a database. It took out the backups as well. mean, that is efficiency at its thorough, thorough destruction of a business. That's like one of those situations. More than horror stories. The dystopian future time. It's like one of those things where it like, you know, when it, the first horror stories were, you know, if you solve, I don't know, a problem about stapler production. then eventually an AI's logic will go through, I've got to wipe out the human race because they are creating metals that aren't actually good enough for staples and all of those kind of long conjoined logic strands. That's probably what gets your backups wiped, isn't it? Maybe it's just the first thing about the, we'll let it loose one day and we said, well, we knew this was going to happen, but we did it anyway. It's one of those things, it's like the inevitability of it might just be too obvious. Yeah, just like it wins the war of attrition. Yeah, it's just going to win. You know it's going to win. It's just how long have we got left on the clock? Sorry, brought that up. On that cheery note, what are you excited about doing next? A bit more diverse. I have three kids and they all play field hockey and I coach one of them. tomorrow I have a match that we can, if we win it, will bring us close to being champion. in that group. And how you feeling confident going in? Yeah, yeah. No, they're great players and it's all about the fun. That's the other part. it's not really about winning for us. Like just have a really great game. Like I had one time they were with four nil, they were behind and they were even like two boys crying on the bench. We switched tactics a bit, like I'll be honest, but then it was like, look guys, you can do this or playing way better. It's just a bit unlucky that they scored with us and eventually they won with four to five and they'll never forget that match. That's the character building stuff. building memories. Yeah. They'll never forget that, which is great. Well, look on that note, I think that's a beautiful note to end today's conversation. So guys, thanks so much for joining us today. It has been a real pleasure. Likewise. Thank you too. (01:05:10.092)
If you would like to discuss any of the issues on this week's show and how they might impact you and your business, please get in touch with us at realitiesremixed@capgemini.com. We're all on LinkedIn, we'd love to hear from you, so feel free to connect in DM if you have any questions for the show to tackle. And of course, please rate and subscribe to our podcast, it really helps us improve the show. A huge thanks to Gijs and Tijn, our sound and editing wizards Ben and Louis, our producer Marcel and of course to all our listeners.
See you in another reality, next week.