Applications
GitLab reports training and usage signals from its AI fluency programme
GitLab describes a federated AI-governance and training model and reports a 22.3% rise in daily internal coding-tool interactions after launch.
GitLab's first-party playbook describes a central Enterprise AI hub setting standards and security guardrails, embedded AI Transformation Owners in business functions, and local AI Champions. Its AI Literacy Ladder provides role-specific training, including engineering workshops for planning, code review, pipeline repair and security remediation.
GitLab reports that 87% of engineering-workshop attendees said they learned something immediately applicable, while 95% of a separate non-technical workshop group said they were likely to apply something within two weeks. One month after the AI Ladders launch, GitLab observed a 22.3% increase in daily interactions with its primary internal AI coding tool. These are feedback and usage measures, not measured engineering quality, revenue or causal productivity gain; the tool is not identified as Claude in this result.
gitlab-claude-enterprise-internal-pilot records Anthropic's separate account of GitLab's Claude pilot. The training programme cannot be used as proof that Claude caused the interaction increase. Related employee-enablement examples include 1password-ai-champions-finance-automation and accenture-copilot-employee-rollout-results.
Source: GitLab internal AI playbook | 2026-09-02