Day 1: Using AI Across the Development Lifecycle

Development Tasks and Boundaries

• Ownership expectations for tasks and boundaries
• Version control across model releases
• Human review of generated code
• Performance thresholds for deployed models

Prompting for Code Assistance

• Application limits around code assistance
• Approval gates before production deployment
• Release records linking changes to evidence
• Regression tests before software integration

Context and Repository Awareness

• Evidence requirements for repository awareness
• Tests covering critical software cases
• Security risks in external dependencies
• Clear requirements prior to code generation

Day 2: Improving Code Quality and Review

AI-Assisted Test Generation

• Performance signals around AI-assisted test generation
• Rollback plans for failed releases
• Monitoring for production model degradation
• Model drift after production deployment

Code Review and Refactoring

• Practical tests for code review
• Trust boundaries around sensitive data
• Data flows across connected services
• Monitoring signals after production release

Security and Dependency Checks

• Validation checks for dependency checks
• Service dependencies mapped before change
• Interface contracts between system components
• Integration failures requiring fallback paths

Day 3: Governing AI-Enabled Engineering Practice

Intellectual Property and Licensing

• Evidence requirements for intellectual property
• Dependencies shaping service reliability
• Capacity limits affecting system performance
• System components and integration boundaries

Human Review and Accountability

• Evidence requirements for human review
• Security controls embedded by design
• Failure recovery for high-impact components
• Workplace telemetry supporting diagnosis

AI Development Workflow

• Application limits around AI development workflow
• Process variation before customer impact
• Verification checks prior to final release
• Sampling evidence supporting quality decisions

• Guided exercise applying code assistance to workplace cases
• Group review evaluating code review against evidence
• Case analysis examining intellectual property in workplace decisions
• Simulation testing AI development workflow under operational pressure