Pythian's AI operating model

In a Google Cloud post written by Pythian executives, the company says it rolled out Gemini Enterprise across a 500-person workforce in 27 countries. For its internal database operations, it describes an agentic workflow that reads tickets, searches knowledge, and drafts runbooks. Pythian reports an 80% reduction in mean time to resolution across a workflow handling about 15,000 monthly tickets, alongside a threefold rise in active user engagement. These are company-reported outcomes; the post does not publish an independent evaluation or comparable baseline methodology.

The post distinguishes Pythian's internal results from anonymised customer examples. Its further claims about customers' IT tickets, supply chains, and retail onboarding should not be presented as independently attributable adoption by those unnamed organisations. The relevant business lesson is the pairing of a prioritised process backlog with production monitoring and ongoing model operations.

Related

• cyberark-bedrock-support-automation — another support-operations deployment with reported resolution-time change.
• regal-rexnord-copilot-agents — employee knowledge and customer-service agent deployment.