AI workloads, especially those involving machine learning models, often have unique and dynamic process behaviors. During training, inference, or continuous learning phases, the processes involved (e.g., data ingestion, model execution, logging, external API calls) can be complex and evolve. Manually baselining these processes is extremely challenging due to: training, complex dependencies, resource intensive operations...

Proposal: Use AI BOM to autogenerate process baseline.