After the platform-fragmentation issue surfaces, Meridian's PM proposes a shared feature store and a common data/model versioning environment so the predictive-maintenance and demand-forecasting teams stop working from disconnected spreadsheets and ad hoc scripts. What is this PM action an example of?
Select an answer to reveal the explanation.
Short Explanation
A shared feature store is just everybody finally working off the same recipe book instead of five different scratch notes.
Full Explanation
Standardizing a shared feature store and common versioning environment across teams is a direct example of creating a cohesive development environment for ML projects — a named Domain III enabler that supports consistency, reduces duplicated effort, and makes collaboration and handoffs between teams (and future audits) far easier than disconnected spreadsheets and ad hoc scripts. Calling this data augmentation is wrong — augmentation refers to techniques that synthetically expand or diversify a training dataset, not to standardizing infrastructure and tooling across teams. Calling this reinforcement learning is wrong and a non-sequitur — there's no agent, environment, or reward signal involved in adopting shared infrastructure. Calling this ensemble modeling is wrong because nothing here combines multiple trained models' predictions — it establishes shared tooling and data practices, which is an infrastructure/process action, not a modeling technique.