Model Versioning
Keeping track of which version of a model and which version of your prompts are running in production — so you can test changes safely and roll back if something breaks. This matters because AI model updates change behavior: something that worked well on Claude 3 might behave differently on Claude 3.5. Similarly, changing a system prompt is effectively a code change that should go through the same review process.
In practice
You hardcode "claude-3-opus-20240229" in your API calls instead of "claude-3-opus-latest." Three months later, Anthropic updates the model. Your app keeps using the exact version you tested, so nothing changes unexpectedly. Model versioning is pinning to a specific version so updates don't silently change your app's behavior.
Related concepts