ModulesSAI-210
SAI-210 table of contents
Model Selection and Lifecycle Management
Evaluate model fit, licensing, provenance, packaging, approval, updates, and retirement inside a controlled lifecycle.
Learning outcomes
What you should be able to do
- Create model acceptance criteria
- Record licensing and provenance evidence
- Define model update and retirement controls
Curriculum
Work through 4 sections in order.
The chapters are individually addressable documentation pages. You can link directly to a concept from another program, architecture decision, or implementation guide.
Lifecycle foundations
Define the model asset, its attack surface, and the lifecycle decisions that require control.
Introduction to model lifecycle management
Manage a model as a versioned governed asset from discovery through retirement.
AI assets and attack surfaces
Inventory the data, models, prompts, artifacts, identities, tools, interfaces, infrastructure, and evidence that require protection.
Artifact identity and provenance
Identify software, models, data, prompts, policy, and configuration with verifiable origin and lineage.
Evaluate and select
Apply explicit fitness, risk, license, and operational acceptance criteria.
Model evaluation, selection, and licensing
Compare model fitness, risk, licensing, operational constraints, and sovereignty consequences.
Evaluation and release gates
Turn acceptance criteria into repeatable promotion decisions with recorded evidence and rollback conditions.
Package and govern
Create an authoritative model record and maintain it through change and retirement.
Model packaging, approval, and registry
Package models with provenance, evaluation, configuration, policy, and approval metadata.
Model monitoring, updates, and retirement
Detect material change, reassess fitness, control updates, preserve rollback, and retire safely.
Controlled change and versioning
Version complete AI behavior, assess change impact, approve promotion, support rollback, and retire superseded assets.
Architecture decision records
Record context, options, control consequences, evidence, dependencies, limitations, and reversal triggers for material decisions.
Evidence and application
Produce traceable lifecycle evidence and test the complete decision.
Evidence by design
Design evidence alongside controls so important decisions, releases, configurations, and operating events can be verified.
Model lifecycle workshop
Evaluate an open-weight model candidate and produce a controlled lifecycle decision package.
SAI-210 knowledge check
Verify model acceptance, licensing, provenance, packaging, release, monitoring, and retirement decisions.
Practical completion package
- Model requirements and acceptance criteria
- Candidate evaluation and license record
- Provenance and model package manifest
- Promotion and rollback decision
- Monitoring, update, and retirement plan
Current release boundary
This public curriculum teaches a model-governance method, not approval of any named model or license interpretation. Legal, security, safety, performance, and workload-specific review remain the responsibility of the adopting organization.