Apollo Global Management, a $900B alternative asset manager, shares how they scaled Temporal workflow orchestration from 100K to 15M+ actions per month across 8 engineering teams. Key pillars included running all execution within their Azure cloud (regulatory requirement), Kubernetes-based workers, Terraform-first infrastructure, and comprehensive observability. They built shared Java/Python libraries wrapping common Temporal activities, automated CI/CD pipelines for workflow deployment, and RBAC-mapped namespaces (160+) for tenant isolation. A notable win: using AI (Claude + OpenAI) to migrate a legacy ETL platform with 100+ jobs and 2M monthly executions to Temporal in just 3 weeks, with 95% of code auto-generated. Their phased adoption strategy — starting with back-office workflows, then expanding to middle and front office — helped avoid the 'boil the ocean' trap and drove genuine consolidation from 7+ workflow systems down toward one.