Coffea-casa analysis facility is one of the components of the Analysis Grand Challenge (co-lead by Alexander Held and Oksana Shadura). This allows the Python packages from AGC pipeline to be tested at scale at the facility even on the stage of release candidates (applying DevOps strategy), making sure we could offer early adopters a reliable development environment for their analysis code.
In this short talk, I will present the evolved architecture of the facility, such as the integration of advanced services (ML, data delivery & data caching, and others) available to end users of the facility.