Worked on the vllm-project/production-stack repository to enhance deployment documentation for KEDA autoscaling by integrating observability monitoring as a Helm sub-chart. This update eliminated the need for a separate observability installation, streamlining the setup process and aligning with the project's monitoring strategy. The work involved updating Markdown documentation and YAML configuration references, including values.yaml and Prometheus service addresses, to reflect the new sub-chart structure. Clarified requirements for prometheus-adapter usage, reducing confusion for new deployments. Addressed reviewer feedback to ensure alignment with v0.1.11+ standards, ultimately improving onboarding, deployment reliability, and operational efficiency for Kubernetes-based production environments.