CI/CD for AI: Automating Docker-Based Model Builds and Deployments Training Course

Introduction to CI/CD for AI Workflows

• Distinct challenges within AI model delivery pipelines
• Contrasting traditional DevOps and MLOps processes
• Essential components of automated model deployment

Containerizing AI Models with Docker

• Creating efficient Dockerfiles for ML inference
• Handling dependencies and model artefacts
• Constructing secure and optimized images

Configuring CI/CD Pipelines

• CI/CD tooling choices and their respective ecosystems
• Developing pipelines for automated model packaging
• Verifying pipelines through automated checks

Testing AI Models in CI

• Automating data integrity validation
• Unit and integration testing for model services
• Performance and regression verification

Automating the Deployment of Docker-Based AI Services

• Releasing AI containers into cloud environments
• Executing blue-green and canary rollouts
• Rollback procedures for unsuccessful deployments

Handling Model Versions and Artefacts

• Utilizing registries for model and container version control
• Tagging, signing, and promoting images
• Coordinating model updates across various services

Monitoring and Observability in AI CI/CD

• Monitoring pipeline and model performance
• Raising alerts for failed builds or model drift
• Tracing inference behaviour across different environments

Scaling CI/CD Pipelines for AI Systems

• Parallelizing builds for large-scale models
• Optimizing compute and storage resources
• Integrating distributed and remote runners

Summary and Next Steps