Cognizant and Google Cloud announced a new group Oct. 19 backed by more than 18,000 Cognizant associates, aiming to help business clients move AI agents from pilot projects into production.
The Gemini Frontier Capability Group will initially work on legacy modernization, developer experience, autonomous agents for business processes and frontier cyber defense. It plans to deploy a standardized architecture for AI agents based on the Model Context Protocol and secured with Google Cloud’s Model Armor.
The companies described the group as a joint investment but disclosed no amount or financial terms. Google Cloud CEO Thomas Kurian said deploying AI agents at scale requires partners with expertise in system architecture and integration.
Cognizant cited an unnamed retailer as an example of its work with Gemini. The company said its integration of Gemini models into the retailer’s software development helped raise developer productivity by 40%, cut time to market by 30% and reduce costs by 35%. Cognizant did not give a baseline or measurement period for those figures.
The new group follows a Gemini Enterprise Practice that Cognizant established in April, when it said it planned to place engineers alongside Google Cloud teams in client environments. In July, Cognizant set separate goals of deploying Gemini Enterprise to 100,000 associates during 2026 and certifying at least 10,000 professionals on the platform.
The companies also said more than 55,000 Cognizant associates have completed a Build with Gemini workshop. That figure measures workshop participation; the more than 18,000 figure describes the associates backing the new group.
Cognizant’s next scheduled company update is its third-quarter 2026 earnings release on Oct. 29 before U.S. markets open. An earnings call is scheduled for 8:30 a.m. ET.
