Bridging the AI Production Gap: SUSE’s Agentic Future

My next guest, though, I want to introduce you to. I've interviewed him before, Abhinav Buri. Yes.

You got it right. Abhinav, first of all, welcome back to Techstrong. Thanks, Alan.

My man, how you been? All good. It's always a pleasure talking to you, Alan.

Yeah. So I imagine it's been quite a show for you as well. Yes.

It's been an amazing time. You've been to a lot of these events. Mm-hmm.

I feel the energy this time here at SUSECON, it's electric. It's amazing. Yeah.

One of the articles I was writing about is an article I wrote called "SUSE Ascendant," where I really feel like this was a transformational event. Well, the event reflects reality. Yeah.

It's a transformational time for SUSE- Yeah ... we're at 30 plus years- Yeah ... of a history, but with consistent open is a theme from day one.

But history, and some people call it luck, but luck is not by design, luck is by hard work. That is true. But the currents of the market and history are swirling at the perfect point.

I'll add to that just one point. Yeah. It's not just hard work.

Luck is where hard work meets opportunity. That's the saying, right? Yeah.

" That's nice. That's a good one. I'll use that next time.

That's from a baseball executive once. Okay. All right.

But anyway- All right ... let's talk a little. You did a keynote.

Was it yesterday? That was yesterday, yeah. Talk to us about your keynote.

Yeah. That's amazing. So here's the thing, right, Alan.

Last three odd years, all of us, each and every person here at SUSECON, forget SUSECON, I mean the people here at this conference center, every one of us has tinkered with AI chatbots one way or the other. We've all used AI features of different software we work with, and now we're all playing with AI agents one way or the other again. Mm-hmm.

And that's happening right now. Yeah. But what you see as you talk to the people here, the CIOs, CTOs, engineers, architects of global platforms, the conversations have fundamentally shifted.

It is no longer about what can AI do, because frankly, what can it not? It is about where is the tangible ROI from my AI investments. Yeah.

And what we are observing is that there is indeed a production gap, so to speak. The gap between experimenting with agents, with chatbots in the last couple of years, to a sovereign, resilient, secure AI ecosystem that can power your AI factories of tomorrow. So that has been the theme of- So there's a mouthful in there.

Let's dissect this- Yeah ... one layer at a time. So first of all, you're right.

AI has been a... Look, I remember when the internet went commercial, and people first- Yeah ... started seeing the web- Yeah ...

and you could see their wheels spin. Yeah. The cloud came out.

Yeah. Frankly, I remember when cell phones- Yeah ... became a thing.

Yeah. Right? They were big.

I was there. Oh, you were there too? You may not believe it.

I was there. Okay. They were bricks, and I'd have them in my car and- Yeah ...

but, and there is an adoption curve. It's the crossing the chasm, 15% early adopters- Exactly ... mainstream, late mainstream, et cetera.

Exactly. The migratory path, if you will, the evolutionary path for AI is first you do chatbots, and it's like a parlor trick. Yeah.

You could do many different types of tricks. Then you start realizing, how does this help me with work? Yeah.

And I think most of us are well into that. Yes. Here at Techstrong for the last month and a half- Yeah ...

we've been on an agentic- Wow, yeah ... experiment. That's amazing, yeah.

As the CEO, I've given everyone orders that they must try it. Yeah. They don't have to like it, but you got to try it.

Yeah. I don't care how many tokens you burn. Mm-hmm.

Because at the end of the day, it's coffee money. And then are you giving them an action to share something with you? Everybody has to report back every week.

Perfect. Yeah. We have a running website that the AI made- Uh-huh ...

of our experiments with a card- Yeah ... for each person's... Most of us have multiple experiments.

Yeah. I have about 26 myself. Wow, that's amazing.

But we have, in a month, I think we have over 80. Uh-huh. And we're a small company.

We're only 25 people. Yeah. So I do think it's there, and I'm going to tell you that it is game changing.

Yes. Game changing. Yes.

As a CEO, I'm telling you- Yes ... it's game changing. Yes.

However, so that's experimentation. Again, part of this evolution is from experimentation to standardization to operationalizing it. Yes.

Now I'm trying to get us to operationalizing it. Yeah. And that's what you talked- Yes ...

about. Indeed. And just like you, when I speak to CEOs, CIOs, CTOs, I think another facet, one thing I hear quite often is: Where are these getting stuck?

When in large scale implementations- Mm-hmm ... because then you need to start looking at the infrastructure- Yeah ... that is powering these end user workflows.

Yep And they're getting stuck because the infrastructure that we have today, that was built to host workloads. That was built to be managed. It wasn't built to think.

No. And that is the pivot that infrastructure software providers like us are bringing to the table. As an example, what we are on a journey on as SUSE, and I'll take a step back before I talk about that journey.

Last 30 plus years, you probably are following and observing SUSE for a while. Mm-hmm. We have made hardware invisible.

We have mastered the art of software defined, but now in 2026- Hardware matters ... that cannot scale. Hardware matters.

Hardware does matters, but humans to pull every single trigger will not scale- No ... today. So hence, we are moving from a managed to an autonomous future- Yeah ...

where we are making our infrastructure a digital agentic collaborator- Mm-hmm ... for the platform teams and the CIOs. " I already uploaded it.

It might be published already, to tell you the truth. Yeah. com.

Platform engineers, you're designing your platform for humans- Yeah ... but the users are going to be agents. Agents.

Yeah, both. Right. We got to start designing our platforms for agentic users.

Absolutely. Not human. Absolutely.

Or, and humans, but mostly agentic. 100% agree with you. And it's easy to say, harder to do.

Yeah. You're absolutely right, and as a CEO or any CXO individual who is architecting that future of their infrastructure, we firmly believe that the strategy should be based on three pillars, okay? Number one is resilience.

And why that is important is, Alan, a few years ago, when we talked about resilience, the conversation naturally would gravitate to uptime. Yeah. Now, that is no longer the case because now resilience is about any organization's ability and capability to absorb shock.

Yeah. Whether that is in the form of geopolitical events, whether that is in the form of technological shifts, or financial. Mm-hmm.

And that resilience also cannot be a reactive process. No. So you need to architect for that because now, in today's age, you cannot patch your way out of a machine speed crisis, okay?

So that's pillar one from a CXO standpoint. Pillar two is about sovereignty. Yeah.

And sovereignty in the agentic era is about control. Because every time now that you build a business process, a workflow around any proprietary API, you are paying a tax, Alan. And it's not just dollars per token, which you spoke about a while back.

That's not the only tax. The tax is about control. Yeah.

Sovereignty, we believe, in the agentic era, is about controlling the entire life cycle of your intelligence- Yeah ... and do you have that? Mm-hmm.

And number three is scale. That's the final one. To be able to do that from your workstations to AI's factories, to far edge- Yeah ...

to thousands of factory floors. So yeah, that's you need to architect for. I don't disagree.

I emphatically agree. To me, it's not just control, it's freedom. Yeah.

" Yeah. " And one of the nice things about AI was having a conversation. Yeah.

It's so polite. I had a thing where it messed up one of the websites we did, and I said, "You messed up this whole website. It was 99% done.

" Mm-hmm. "I'm so sorry, Alan. " Oh, okay.

Yeah. I love it. I wish every workers were like that.

Yeah. But polite's not enough. Yeah.

The nice thing about it is you've got, you want to call them digital workers? Yeah. Digital workers who are so capable, who will follow instruction.

Yeah. Who will do- Yeah ... what you ask them to do.

But it's only as good as your ability to make those choices. Yeah. Control is one way.

Yeah. Control to me sounds like I control you, right? Yeah.

What I think of is I've got an open road. Yeah. And when I come to that fork- Yeah ...

I decide- Yeah ... do I want to go right, left, or in the middle? Yeah.

Right? And that is, I guess, control as a way, but it's freedom. So the angle for control is on the intelligence angle, and I'll clarify that a bit.

So here's the thing. When these language models or these AI agents are interfacing with your proprietary workflows and business processes and your data- Mm-hmm ... they are effectively encroaching on, one, your data.

Yeah. Two, potentially your IP. Third, your core business processes.

Yeah. And one, two, and three together, in most cases, contribute to your organization's competitive moat. Yeah.

And control in this sense is about controlling the life cycle of your intelligence. Yeah. And freedom comes from the ability to decide your future.

To make that decide. Exactly. And that's another factor.

You're following SUSE, we stand for choice. Right. That's a strategic principle for everything we build.

Yeah. And I think that's extremely powerful, especially in today's age. Absolutely.

The other thingAnd things I learned at SUSE CON. Uh-huh. Right?

Resilience is tied into this sovereignty- Yes ... issue. Sovereignty is not even...

There's different levels of sovereignty. There's national sovereignty. Yeah.

There's organizational sovereignty. I think another thing, and we haven't really touched on it yet, but believe me, it's going to be important. Mm-hmm.

Individual sovereignty. Mm-hmm. Right?

You get to control your personal choices, your data, your processes. Yeah. Your financials.

Yeah. Yeah. You know what I mean?

And I think that's going to be yet another turn of the wheel- Yeah ... in when we talk about AI and resilience and because all of us... I was talking, I forgot, Andy Fitzsimon.

Uh-huh. I was talking to him. And he said everyone here, not just the people at SUSE, the people- Yeah ...

I don't know if they could see, but the people behind the counter here at the Hilton- Yeah. Yeah ... they've all played with chatbots.

Everyone. Yeah. Everyone is trying it.

Yeah. The ability, though, to use it. Mm-hmm.

The ability to decide how involved you want it to be. Yeah. Right?

Because we've all thought about: What does it mean for my job? Yeah. What does it mean for my career?

What does it mean- Yeah ... for the ability for me to feed my family, to provide for my loved ones? Yeah.

And I do have a take on that, Alan. Go ahead. So let's be real, right?

It is going to impact certain roles and certain jobs. Right. Well- However, this is going to evolve, and it will evolve into the requirement for a different type of skill set.

Yep. So while certain jobs and certain skills and roles will get impacted, a large new set of jobs and roles will also get created. Yeah.

And that is the future we are headed towards, and those are the skills that will be most important. We are here- It's the future right now ... living in.

Yeah. Yeah. Tip of the spear with software developers.

Exactly. Today, software developers are not coding. Yeah.

They're managing their AIs to code. Yeah. And that's a different job.

Exactly. A different job skill. And those coders who say, "Oh, no, I'm going to do it by hand," they don't scale.

They're going to have a hard time finding a job. I've spoken to so many peers of mine who are C-levels at bigger companies than me. The first thing they ask a developer today on a job interview, "What AI do you use?

" So coding, software engineering is already changed. Yeah. Tech marketing, product marketing, marketing comms, if you're not using this today- Yes ...

I'm sorry, but I don't know if I'd want you on my team. Yeah, and you make an interesting point. So it is not about that replacing individuals.

It's not about you being replaced by an AI. It's about you being replaced by a person who can leverage it better than you. Who can do it.

And, you know? Yeah. So it's about upskilling.

Right. That's the whole argument. You hit it.

And we spoke earlier about my experiment here, telling everyone to use- Yeah ... that's the whole point. I'm not looking to replace you.

I want to 10X you. Correct. I want you to be 10X the person, the employee you are today.

Exactly. Because I don't have the money to hire 10X more people. Yeah.

But I can get 10X hiring using agents with you. Exactly. And that, as a CEO, as an entrepreneur- Yeah ...

that's what excites me. That's very powerful because that's also what we hear from most CEOs and CIOs we work with. The budgets are not increasing.

No. If anything, they are decreasing or stagnant. Yeah.

But the concept of doing more with less is even more pervasive now than it was a few years ago. That's- And how do you do that? need to do.

And now translate that back to the infrastructure story and how we as a company, as SUSE, are helping our customers in this space. And because that's also our focus, we are following a two-pronged approach across our portfolio to help our customers in this era. Number one is what we're calling AI for infrastructure.

Yep. Where we are infusing AI capabilities across our portfolio. Yeah.

From cloud native to Linux, everywhere. A to Z. Exactly.

To automate a lot of complex operations and free up a lot of capacity for the platform teams to focus on innovation. Yep. So that is one part of our play.

That's part of the 10X. That's part of the 10X thing that you mentioned. Yeah.

The part B of our play is infrastructure for AI, providing a turnkey platform for our customers to run sovereign, private enterprise AI implementations. Yeah. And that is also going to be key.

Because here's another angle we can possibly discuss, Alan. All of these agents that we're talking about, the actions they're going to commit, at the end of the day, they are non-deterministic. Yeah.

So how do you, as an IT leader, ensure that whatever these agents do, they are running and they are empowered to act, but they're not running wild? Wild. So putting deterministic guardrails around these non-deterministic agents is a very important conversation point for today.

So that's the governance aspect of it. The governance aspect. And that's where we are today.

Look, as we spoke, I've seen other tech waves. Nothing as big as this wave, but I've seen other tech waves, and this is the way it goes, is first you have just wild experimentation. A lot of it, in hindsight, is wasteful.

But you separate the wheat from the chaff, so to speak. The next step is the governance phase. Yes.

Because that's where stuff gets real. That's where it hits the fan. Today.

And I think that's where we're going to be, but the thing about this AI wave is what maybe that used to take a year or two- Yeah ... it's probably going to take a few months at most. It'll be much faster.

It'll be much faster. And that's what we're dealing with. That's what we're living in.

What kind of feedback did you get on this? So the feedback is incredible. So we have customers here who are actually running digital twins of their entire factories, which are basically AI simulated twins of their factory setup, where they're running simulations on any changes in the twin before they actually implement it in their hard production setups.

And that is a reality today. It is. I agree.

And I think digital twinning, we've just scratched the surface. That's going to be something. Mark my words.

Yeah. We get together next year for SUSE CON, we're going to be talking- That's all we'll be talking about ... a lot about it.

We'll have hundreds of customers doing that with SUSE. Absolutely. They're giving me the sign here that people are waiting.

All right, Alan, it was always great talking to you. Time just flies. A pleasure, man.

Hey. Time just flies. We're here at SUSE CON.

We've got a lot more coming your way. Actually, we're going to be back on in about a minute.