NEW YORK – The word "evolve" was prominently displayed throughout the Marriott Marquis during Cloudera's annual user conference, held at the hotel in the heart of Manhattan's Times Square.
Evolve is the name of Cloudera's event. But more than just a title, it's a message.
In the burgeoning AI era, data management vendors need to adapt to a new way of doing business to remain relevant. Founded in 2008, Cloudera was one of the darlings of the big data era. But when the cloud emerged, Cloudera -- perhaps ironically, given its name -- was slow to augment its on-premises capabilities with cloud-native tools and was usurped by upstarts such as Databricks and Snowflake along with hyperscale providers AWS, Google Cloud and Microsoft.
AI is giving Cloudera a chance at a comeback. And the company appears to be taking advantage of the opportunity, evolving to catch up with current trends as well as finding a way to distinguish itself.
Like its peers, Cloudera is providing capabilities such as AI Agent Studio aimed at enabling customers to build AI agents that can be trusted in production. Somewhat uniquely, however, Cloudera is enabling deployment across cloud, on-premises and hybrid environments.
Rising concerns related to regulatory compliance, data security, data sovereignty and the cost of running AI in the cloud is making on-premises and hybrid deployments an appealing option. Along with Teradata, Cloudera is distinguishing itself from the rest of the field by providing such capabilities.
"The on-premises capabilities is the clear differentiator between those two and the [competition]," David Menninger, an analyst at ISG told TechTarget. "In a funny way, the market has come around to where Cloudera is. … Now, hybrid is a valuable capability."
Boaz Rubin, head of network software R&D at Cloudera customer Amdocs, similarly noted that not only is Cloudera offering a needed option by enabling hybrid and on-premises AI deployments, but it is understanding what users are going to need as they build networks of agents and other AI tools.
"Cloudera is going in the right direction," he told TechTarget. "They're trying to be the bedrock of on premises, [which] is very exciting because a lot of [companies] are starting to realize that the cloud is very expensive. ... I believe that the whole world will come back to on premises more and more."
Evolving to meet customers' needs
During the keynote address at Evolve on Oct. 27, Cloudera revealed that it launched a set of prebuilt AI development packages called Cloudera Blueprints. The Blueprints, so far a set of four industry-specific blueprints for crafting AI agents -- but with plans to add more -- are designed to help customers more successfully build and deploy agents.
Making the development accelerators generally available follows closely on the August unveiling of Cloudera Anywhere Cloud, a platform that will enable users to build, deploy and scale data and AI applications across any environment, including on premises. General availability is expected by yearend 2026.
In addition, in September, Cloudera formed a partnership with AI model provider Mistral to make Mistral's models natively available in Cloudera's platform so users can bring AI to their data without incurring the cost of moving data between environments and potentially exposing data as it travels between environments.
Cloudera is going in the right direction. They're trying to be the bedrock of on premises, [which] is very exciting because a lot of [companies] are starting to realize that the cloud is very expensive. ... I believe that the whole world will come back to on premises more and more.
Boaz Rubin
Head of network software R&D, Amdocs
Capabilities such as Cloudera Blueprints, Anywhere Cloud, native integration with models and AI Agent Studio -- which is part of the Cloudera AI Studios suite of tools launched in August 2025 -- demonstrate that Cloudera is evolving to become a modern data and AI provider, according to Sanjeev Mohan, founder and principal of analyst firm SanjMo.
"Anywhere Cloud is breaking through the notion that Cloudera has been built on layers and layers of [outdated] Hadoop technology," he said, noting that the new platform is built on the open-source Apache Polaris catalog and features the open-source Trino query engine to power massive workloads spread across different data and AI environments.
"It is a completely rearchitected from scratch ecosystem," Mohan continued.
But more than demonstrating that Cloudera has evolved, the vendor's new capabilities are addressing growing enterprise needs ahead of some competitors.
Many enterprises in industries such as healthcare, financial services and telecommunications that handle sensitive information have been slower to put agents into production due to the security risks autonomously acting AI tools pose. Meanwhile, concerns over data sovereignty are on the rise as is apprehension over the high cost of developing and managing fleets of agents in the cloud.
On-premises, private cloud and hybrid deployments mitigate each of those concerns by enabling enterprises to avoid the cloud.
"The recent Mistral capabilities that let us keep our data within our networks yet use some of the advanced AI features are going to be a very powerful thing for us," Paddu Vedam, chief innovation and modernization officer at Cloudera customer Gainwell Technologies, told TechTarget. "That is something we are very eagerly looking forward to."
Cloudera: ahead in one area, playing catch-up in others
Although Cloudera is offering some of the same tools as its peers, and despite it finding a way to stand apart by providing capabilities that enable on-premises and hybrid AI deployments, the vendor has to further evolve to fully match the breadth of other platform providers.
Menninger noted that Cloudera remains good at serving large companies in highly regulated industries with complicated businesses. What it does not do well is provide capabilities that organizations without deep data and AI engineering teams can use for their needs.
"What they have to do better is the quick, easy startup, letting someone turn something on and have immediate access," Menninger said. "They've been heading in that direction, but they can probably go further."
Mohan noted that Anywhere Cloud will be a significant addition when GA, making Cloudera's platform more current and competitive.
"There is an opportunity for Cloudera with Cloud Anywhere to keep pace with other vendors," he said. "Now it has an open foundation to introduce a decision model, for example. Cloudera is at the cusp of making the status quo, if not innovating."
However, changing the perception that it is a vendor that thrived in a bygone era will be a challenge for Cloudera, Mohan continued. But it's not impossible. Microsoft was originally known for Windows and initially fell behind AWS when the cloud emerged, but rebounded as a leader in the cloud witg Azure. Oracle, known as a database specialist, has similarly changed how it is perceived and is viewed as a cloud infrastructure, data platform and AI development provider.
"The biggest issue for Cloudera is to break that mold," Mohan said. "In the early days of Big Data, Cloudera was the foremost name. Then others took the lead with the cloud, but now with hybrid, the market has come to them. … If Microsoft and Oracle can change their perception, what's stopping Cloudera?"
As a customer, Rubin would like to see Cloudera evolve in certain ways as well.
He suggested that while it's vital for Cloudera to add the capabilities competing vendors are providing, it can't do so at the expense of core capabilities users rely on such as the database Amdocs and many other Cloudera customers use as part of their data and AI workflows.
"Their focus is on new things and how to grow in different areas, but I have more issue with the database that has not evolved so much, that they haven't made containerized like MongoDB," Rubin said. "Everyone has moved to containers, and they're still a bit behind. ... But the company is very open, out to bring new technologies to on premises, which is a direction I've not seen from any other vendor."
In addition, he noted that the AI development capabilities Cloudera is providing are somewhat rudimentary and that more developer-oriented tools are needed.
"If they were able to bring it to the next level for [research and development], for developers, and add more open-source [compatibility] to be more flexible, then the sky is the limit," Rubin said.
Eric Avidon is a senior news writer for Informa TechTarget and a journalist with more than three decades of experience. He covers analytics and data management.
