Stage 2 · Lesson 5 of 6
Model registries and the inventory
An inventory lists systems; a tracks the models inside them. A registry, such as the one in or a cloud ML platform, stores each model version with its training data reference, code version, parameters, evaluation results, who approved it and which stage it is in (staging, production, archived).
Linking the inventory entry to registry versions lets you answer "which model made this decision last March?", which audits and incident reviews ask. For models bought through an , record the provider, the model name and version, and the date you checked their documentation, because providers update models behind the same name.
Common trapa registry that only the data science team can read, while the inventory lives in a spreadsheet nobody links to it.
A question to check yourself, an AI practice session and 1 cheat sheet entry come with the full roadmap. Unlock
Resources for this stage
Start with· follow one of these
- GuideFreeNIST AI Risk Management Framework (AI RMF)
National Institute of Standards and Technology (NIST) · Intermediate
The Map function is the reasoning behind every inventory entry: context, purpose and who is affected
- GuideFreeEU AI Act: regulatory framework for AI (European Commission)
European Commission · Beginner
Provider, deployer and the risk tiers, the two labels every inventory entry needs