Cognizant and Google Cloud announced an expanded partnership on October 19, 2026, and created the Gemini Frontier Capability Group to help global enterprises move AI agents from pilot projects toward production. The announcement describes more than 18,000 dedicated Cognizant associates for the group. Cognizant also reports three business gains for one unnamed retailer after Gemini models were integrated across its software-development lifecycle.
NeoTeo previously covered Google’s Gemini work-agent announcement. The October 19 partnership adds a separate services effort focused on enterprise agent projects.
What the group plans to build
The companies plan to focus the group on legacy modernization, developer experience, autonomous business-process automation and cyber defense. Their proposed standardized, security-first agent architecture names Model Context Protocol (MCP) and Google Cloud Model Armor.
The planned solutions span three industries: a Catalog Support Agent for retail, a Neuro® San-orchestrated Healthcare Agentic Marketplace, and multi-agent Order-to-Cash workflows for manufacturing. These examples describe the initiative’s intended work, alongside its broader aim of moving enterprise agents from pilots toward production.
Cognizant said more than 55,000 associates completed its Build with Gemini workshop, where participants built agents and applications and earned Gemini skills credentials. In July 2026, the company said it was deploying Gemini internally to 100,000 associates during 2026, with plans to scale to 200,000.
The retailer outcome Cognizant reports
Cognizant says one unnamed retailer recorded the following results after Gemini models were integrated across its software-development lifecycle:
| Measure | Change Cognizant reported |
| Developer productivity | 40% increase |
| Time-to-market | 30% reduction |
| Costs | 35% savings |
