Why Nutanix Is Becoming A Platform Company And Its AI Roadmap: CEO Rajiv Ramaswami

Nutanix’s CEO said there are several key pillars to his company’s evolution in transforming into a platform company, including expanding its partner and vendor ecosystems, innovation stack and routes to market.

Rajiv Ramaswami is transforming his best-of-breed hyperconverged company into a platform provider aimed at becoming customers’ “de facto platform” where they can run their applications and manage their data regardless of where it resides.

What does becoming a platform company mean for Nutanix?

First of all, for customers, it means that a platform has many use cases. It becomes much more central for customers in terms of relying on Nutanix to run all their mission-critical applications.

So being a platform provider means now they can run all their applications to serve multiple use cases, not just virtual desktops, but all their mission-critical applications, their existing applications, their newer applications with containers and things such as generative AI.

How else is Nutanix transitioning to become a platform provider?

A platform company needs a broader ecosystem. So our ecosystem is growing fast for us.

For example, [key is] what we’re doing now with Pure Storage, Dell, Omnissa, Nvidia, etc. At our Nutanix Next event two years ago, we had about 25 sponsors from the ecosystem. This year at Nutanix Next 2025, we had 86 sponsors. They’re the who’s who of the industry.

That is, again, recognition of the fact that if you’re a platform company, you’re going to have a broad set of people who interoperate, work on top of and work with the platform.

What other areas is Nutanix investing in to become a platform company?

The third is about route to market. As a platform company, you’ve sort of reached a level of significance where now you have broad routes to market.

Our primary routes to market have always been through our value-added resellers and distributors. That continues to be the way, but you’ve also seen now that we’ve added a couple of additional routes. Over the last few years, we now have some strategic OEMs taking us to market: Cisco, Dell, Lenovo, etc.

They also go to market through their channel partners. So their channel partners can now participate with us, either directly working with us through distributors or they can also work through some of our OEM partners.

The more recent one has been our thrust around opening up managed service providers or cloud service providers to provide broader access. So we have a broad set of go-to-market actions, all relying on partners 100 percent, but just making our platform available to the broadest group of customers in the way they would like to consume it.

The fourth piece is that we are now also available on the online cloud marketplaces. So AWS and Azure, and soon, the Google Cloud marketplace.

What is Nutanix’s AI strategy?

There are two parts to AI. There is training of foundational models, which largely is going to be done by a few large companies that is a very expensive process.

But then the bulk of the customers are going to take those pretrained foundation models and use them inside for real-life applications. So that’s called ‘inferencing,’ and inferencing is where Nutanix provides a great platform.

Some concrete examples of inferencing is there’s a large bank in Hong Kong that’s using Nutanix. What they do is they record all the phone calls that their salespeople make to their customers, and behind them they have a proprietary application that then summarizes the nature of those phone calls and also looks for patterns of noncompliance. That’s one example of an inferencing application. It’s got a built-in pre-trained LLM in there that they’re using. Now these types of applications are growing in the enterprise, and they will need to run wherever the data is present.

So why should customers place their AI bets on Nutanix?

What we have enabled with Nutanix AI is on top of our platform—that can run on any underlying Kubernetes stack—instead of AI inferencing capabilities, it’s a full stack for running agentic AI applications in a very simple manner.

From a customer’s perspective, now you have an infrastructure that you don’t have to worry about that delivers a turnkey AI inferencing stack. So that they can simply focus on building their AI applications and running them on the platform.

So if you look at the stack itself, we have a virtual machine stack, we have a Kubernetes stack, we have a AI agentic stack and on top of it resides the AI application.

So it’s a very simple way for companies to build out their inferencing applications and run them on a platform without having to worry about all the details of what it takes to go build that platform and put it together in the first place.