Hi @square-wire-39606 @dry-beach-38304 👋 I’ve been preparing for the Kubernetes CI/CD integration project and wanted to get your guidance on a few things. So far I have: • Set up Metaflow locally using WSL • Explored the codebase and Runner API • Run the unit test suite (~97 tests passing) • Investigated some Kubernetes-related test failures and tried to understand their behavior I’m currently trying to understand how to integrate the Metaflow Dev Stack (Kind + Argo) into a CI workflow using GitHub Actions. Before proceeding further, I wanted to clarify: 1.Is Python 3.10 the expected baseline for running the test suite, especially for Kubernetes-related tests? 2.For the CI workflow, should the dev stack be fully initialized during each run, or is there an expectation to optimize via caching/prebuilt setups? 3 .Are there existing test markers for backend-specific execution, or should defining them be part of the project? Would really appreciate your guidance on aligning my approach with the project expectations. Thank you