Name: The Great Restructuring: Rethinking Talent Strategy in the Age of AI
Uploaded: 2026-06-01
Duration: 16 min 45 s

The org structures and roles that today’s companies rely on won’t get them where they need to go next. As AI, an evolving consumer, and shifting market dynamics redefine how work gets done, leaders must rethink job titles, career pathways, and value creation. In this fireside chat, Cognizant CEO Ravi Kumar explores emerging interdisciplinary skills in an AI era, how jobs are evolving, and how to redesign work to amplify human potential alongside AI, turning workforce transformation into a competitive advantage. 

Ravi Kumar, Chief Executive Officer, Cognizant
In conversation with: Moderator: Diane Brady, Executive Editorial Director, Fortune; Co-Chair, Fortune COO Summit 

The Great Restructuring: Rethinking Talent Strategy in the Age of AI

Conferences

I want to start with what I read this morning, which is that you've created two new job categories frontier certified Engineer and Frontier Business operator. So I have to ask, sir, what is that and am I eligible to apply? What are these new jobs you've created? You always have excitement when there's a new technology which is disrupting us, so thank you for the invite. Um, you know, I, I used to be a chief operating officer before, so I'm so excited to see a community you've established. So thank you for that. It's also one of the most undervalued. Jobs. Um, you said widespread agreement. Look, yes, yes. You know, did you do it? Nobody thanks you. When you're not around, a lot of people look out for you and they start to ask for work, you know, what, what's going on. So, you know, actually till 2007. 58% of the Fortune 500 did not have a chief operating officer. And I believe it's the most important role in any company. It kind of sits in the hustle of the company. OK, so coming back to your question, no, but, but it's, it's related because it's about how jobs change and, you know, the function of the COO, and now you've created these new functional roles for the age of AI. Yeah, I mean, you know, we, we lived in an era where Intelligence was kind of a premium. Expertise was a premium. The deeper you go into a job, uh, you get paid more, and that was the asymmetry between enterprises and the symmetry between individuals. We're now getting into an era, or rather we're in an era where intelligence is going to be on our fingertips. It's probably going to be at a throwaway price. It's not gonna be the symmetry. How you apply it is going to be the symmetry. So the two roles we have established are roles where they sit at the intersection of technology and a domain or technology and operations, uh, technology and the domain because this is software engineering, it is contextual, it's not deterministic like the classical software we wrote so far. It's contextual, so it's a different science. So we need lean, leaner 1% teams, 1% parts or squads as we call it, which has intersection of technology and domain. It could be a history major with skills to identify and use agentic work. It could be a biology major who does life sciences. It could be a chartered accountant who can use agentic cloud terminals around them and actually deliver work. The business operator is like a BPO plus plus plus. Once you integrate the workflows with, with digital labor and human labor, once you've reinvented those flows, somebody has to operate them. And that is the frontier operator. We have in a loosely packed way used the word forward deployed engineers in the last one year. I mean, forward deployed engineers were built for an era which was very different. Forward deployed engineers were built. It was used by Palantir. Palantir built this because they had a machine learning ontology, which was a black box. They got an engineer. The worldview was everything is broken in a company we're going to bring it together. That's not the worldview today. The worldview is how do I integrate digital labor into the flows of a company. That's why we build these two roads. Well, and when you and I met um probably about a year and a half ago, we talked, and there's been a lot of talk since about these flattenings of the pyramid, the Hollywood studio model. Let's start with how you. Hire and who you hire you mentioned biologists, history majors makes me think, first of all, what business are you in and is that changing, but the entry level worker is the worker that traditionally we look at now is not getting those jobs. Talk about how you see that entry level worker. Yeah, so our universe as a company is no longer software engineering. We want to embed. Into the operations of companies digital labor as they call it, which is gentic in nature, so the universe goes from just tech span of enterprises to oppan of enterprises, um, you know, way back in last year, you and I wrote something together. We believed that the pyramids of work will be flatter. In fact, people said there could be a diamond, there could be a reverse pyramid. We actually felt you need more early careers because you could create significantly more throughput at the bottom of the pyramid. In fact, the entry barriers to new jobs is significantly lower. So you, you create much broader pyramids and shorter pyramids. So the, so the path to expertise is much shorter. The number of people you could hire at the bottom of the pyramid is much higher. So this is counter, it's a counter narrative to what everybody was saying so far, and it's actually a counter narrative also to people who actually said there'll be lesser number of jobs in the future. In fact, in the last one week, the narrative has changed again. Everybody now thinks we need more jobs. There'll be more jobs done by, yeah, in the last one week. Why, why in the last one week? Because now, because now everybody is seeing it. I mean Sam Altman said what I said a year ago is no longer true. I mean, if there was a little bit of fear mongering, drum beating about the fact that there's going to be a collapse of jobs, and I think they're gonna be more jobs. AI is gonna be in the middle of the floor. You're gonna have a ton of jobs in the front. You, you'll have a ton of jobs in the back. These are gonna be validation and verification jobs, and those are gonna be authentication jobs. Now when you have a flatter pyramid, the biggest challenge is. The middle layers. are going to be leaner. So with 15 smaller teams. You're going to have player coaches. You're not going to have people who are coordinating. And pushing information up and down and reducing the symmetry. You're not gonna have them. You're gonna have a broader pyramid with player coaches in the middle. Then how do the people at the bottom. Get heritage of companies. How do they learn? How do they walk the corridors, which is what we did learning in companies. The nodes in a company are going to be more real-time. And those nodes are going to be more genic and that's the Hollywood model so which essentially means if these nodes are more real time and more agentic and you have people who are player coaches, you can actually build harnesses around the tokenization. And you could capture the digital, you could capture the digital labor because it leaves traces. And that has to be your new learning opportunities. So people who walk the corridors have to now watch for people. Who are like them and watch for digital labor which is trapped in the workings of the company and learn out of it and that's the only way you could actually build expertise because if you're doing validation and verification jobs you need experience. That experience is not going to come from just human labor, it's actually gonna come from human and digital labor. So Ravi, I want you talked about this audience and it's, you know, when you're in an operating role you have tough choices, you know, to spark joy. You have to take some pain. You have to, you know, to reinvent. You have to not do the things you did before. So you just put aside up to $270 million for restructuring, layoffs, and yet you're, you said you're hiring more people. Talk a little bit about that process and those choices that you have to make. Look, you know, we're broadening the pyramid. Which means I'm gonna hire more entry level school graduates than last year. Last year we hired 20,000. This year we're gonna have more than 20,000. The year before we hired 12,000 school graduates. So as you broaden the pyramid and restructure, you're hiring more entry level. You're hiring more non-STEM disciplines. I spoke about historians. I spoke about biologists. We only hired engineers before. So we are hiring non-STEM disciplines. We're hiring player coaches. What we want to do is to give people a chance. Wherever they're actually working on coordination orchestration kind of jobs, we have something called an AI fluency leader in the company we kind of ask people to navigate through the process and become player coaches. I mean player coaches higher productivity, having small teams, interdisciplinary talent is what the future is gonna look like and as long as you are in that mix. Uh, you're gonna, you're gonna do well. We will hire significantly more number of people than last year, but we're going to hire broader at the bottom, and we're gonna hire interdisciplinary talent. So you mentioned tokens. I want to get to a headline you had recently that it is not billable hours or a measure of value. What are we getting wrong in your mind about tokens and how that's impacting the way we do business? I mean, for the last two years. How you consumed tokens? How much tokens you consume? was a vanity metric. I think it still is. I mean, people on a drink are talking about, oh, I've spent $500 this month on tokens. Somebody else has spent $1000 on tokens, and we incentivize people. I actually know about a company which has offer letters with token amounts attached to in the offer now. It doesn't tell you what's the outcome, what's the value. So, I don't think you should equate this to number of billing hours. I don't think you should equate this to productivity. So we believe. Tokenization is going to be a science. It's a contextual science. This is a contextual, uh, you know, technology. It has to be grounded in the company's hustle. In fact, that's why the chief operating officer is so important. The capability of this, of the, of the technologies is right out there. The production value of the technology is here. And the bridge to that production value is what is going to make companies successful. That bridge is going to come from grounding the technology into the rules of the company, grounding the technology into the hustle of the company. In fact, we have a science called context engineering. I've actually written two HBR articles on it. If you get a chance, you should read it. I've actually said how to teach AI to be a part of your team. Brewing coffee in Starbucks is very different to brewing coffee elsewhere. Now, the deterministic parts of The company has already been absorbed by the classical software, which we wrote for the last 50 years. The probabilistic part of work is what we're trying to capture. And that context is inside companies. It has to be embedded into the companies. It has to be embedded into the agentic work you do and the tribal knowledge, the, the, the context, the heritage, the culture, it is, it has to be captured in some form, and you have to apply that to digital labor in a way. You have to reinvent the flows and integrate that back in. You know, one of my favorite examples here is autonomous cars. The technology was smart. Back in 2006. We are in 2026. 0.5% of the car rides in the United States are on autonomous cars, because we're trying to ride it on infrastructure which was built for humans to make it more complex, it's humans and machines working together. So, The reinvention of flows in a company. Integrating this digital labor, which is, you know, it's, it's, uh, you know, it is very jagged. It does things which are not as good as humans. It does things better than humans in some cases. So how do you put all of that in place? I think is a lot of heavy lift. We are underestimating that heavy lift. So I want to get, uh, in terms of the philosophy of you as a leader, that Peter Drucker question of what business you're in. Let's go back to imagining what business is cognizant in today and if we're sitting here this time next year, pick a timeline, what business will you be in? Because I think these new tools change how you go to market, they change your relationship with customers. How are you thinking about the business? So we, you know, look, every company, every industry, was born for a purpose. It kind of addressed the white space. The purpose we were built for is to deliver expertise. When there was a disruption. And the way we delivered that expertise was. You know, to be honest, billable hours. I mean we did fixed price work we did outcome based work, but the real thing was billable us software companies. Delivered the value on per seat models and we delivered the value on billable hours. We have to reforge our first principles. For the first time after many, many years we get a chance to build this as bespoke. I mean, we were system builders. We went to system integrators. Now we are bespoke AI builders. We went from delivering on a pyramid with deep expertise to delivering interdisciplinary skills. That's a science we have to build. It's a contextual science because this is a technology which is probabilistic. So you can build that experience, you know, I actually strongly believe we have built a harness on tokenization. You flow the tokens through companies like us, we should be able to deliver better than you delivering digital labor on your own. So going from Deep expertise to interdisciplinary skills. We were a services company. Our clients bought infrastructure. They bought software. They bought services. They owned the agency of outcomes. Clients are now saying, wait a minute. I mean enterprises are saying, why are we owning the outcomes? We've got a company to come and own outcomes for us, so we should be able to own the outcomes we have to go from delivering projects. Delivering billable arts to owning outcomes. And finally, we have to underwrite those outcomes and be paid for those outcomes. I think that is the future. I believe we are going to reforge and whoever reforges this future is going to be a winner on the other side. And very briefly, advice to your fellow, I know you're a former CEO. What advice do you have to COOs? I mean, you know, the bridge to enterprise value is huge. The capability is going up and up. And the production value is right out there. And chief operating officers, chief technology officers, and there's a lot of, a lot of people who do dual roles, they play a big role in bridging this gap. I mean, I sit on two public boards. You know, when I'm on a board, you're pushing all these executives saying, oh, there's no AI in this company. And when I flip and I become an executive, my board is actually pushing me and saying the same thing to me. So the capability is right out there. Everybody is reading it, but the real value of production value is much lower. The velocity to bridge the gap is actually, you know, it is urgent. And the reason it's urgent, unlike in the past, every time there was a disruption, railroad disruption, You go to the digital disruption, the Internet, we got years and years and years to get there. In this case, the scaling loss for this technology is only 6 months, which means, which means, if you can't, if you can't actually get the value by then this is going to go obsolete. So there is an urgency for the velocity to be bridged. The velocity of technology capability is very high. And the production value is here I think chief operating officers are probably the most important people to bridge that gap and if we can't bridge that gap, this will be a bubble because it'll be an industrial bubble it'll, you know, you will have an extraordinary technology, but you're not gonna get value out of it.