Artifacts & Registry

What a model artifact contains, the registry lifecycle from candidate to archived, and promotion judged on quality, latency, memory, cost and robustness rather than one offline score.

What a Model Artifact Contains
▶ lab

A weights file alone is not a model. The artifact is parameters, architecture, fitted preprocessing, feature order, version and training metadata — and serving needs all of it.

Q · The training run finished and produced a file. What has to travel with that file for a serving process to reproduce the predictions the evaluation measured?
The Model Registry

The registry is a state machine over artifacts — Candidate, Registry, Staging, Production, Archived — that stores lineage, metrics, approvals and the feature-definition version, so "which model is live" has one answer.

Q · Three teams retrain the same model on different schedules. How does anyone know which artifact is in production, what it was trained on, and who approved it?
Promotion Is a Checklist, Not a Score
▶ lab

A challenger is promoted on quality, latency, memory, cost, robustness and — where relevant — fairness, compared against the champion on the same slice with the same threshold policy. One improved offline number is not a reason to ship.

Q · The retrained model has a better validation score than the one in production. What else has to be true before it replaces the champion?
Artifact Integrity

Hashes and signatures prove the bytes serving loads are the bytes that were evaluated. Deserialisation formats that execute code on load, and a serving process that loads the wrong file, are the two ways that proof gets skipped.

Q · Between the registry and the serving process the artifact is copied, cached and loaded. How do you know the file serving opened is the file that was promoted — and that opening it is safe?
Preprocessing Lives in the Artifact
▶ lab

The normaliser's means and standard deviations, the encoder's vocabulary, the imputation values, the feature order and the threshold are fitted on the training fold and ship with the weights. Recomputing any of them at serving time is a different model.

Q · The serving path has to turn a raw record into the tensor the model expects. Where do the numbers that transformation needs come from, and what happens if it computes them itself?