Highlights
- Understand the MLOps lifecycle
- Version models and datasets
- Package a model as a container
- Build automated CI/CD for models
- Implement canary deployments
- Configure rollback triggers
- Manage dependencies reproducibly
- Set up a model registry with gates
- Monitor deployment pipeline health
- Document for compliance review
Course Details
MLOps lifecycle walkthrough:
every stage from experiment to production to deprecation illustrated with real pipeline architecture examples
Model versioning lab:
tagging models with metadata, linking them to their training datasets, and querying full version history
Containerisation workshop:
Dockerising a model inference server, writing a health endpoint, and running integration tests locally
CI/CD pipeline build:
automated test, build, and deploy triggered on merge using GitHub Actions or Azure DevOps with working configuration
Canary deployment lab:
routing a configurable percentage of live traffic to a new model version and monitoring quality across the split
Health check and rollback:
defining numeric failure thresholds and testing automatic rollback behaviour in a staging environment
Dependency management:
pinning library versions, building fully reproducible environments, and resolving the conflicts that arise
Model registry setup:
configuring approval gates, promotion stages, and a complete audit trail using MLflow in a team environment
Pipeline monitoring:
alerting on build failures, deployment errors, and latency regressions before they affect production users
Compliance documentation:
producing a structured deployment record and change log suitable for an internal security or audit review
Who should attend
IT and DevOps
Feedback
4.8 out of 5 average
"Our tailored course provided a well rounded introduction and also covered some intermediate level topics that we needed to know. Clive gave us some best practice ideas and tips to take away. Fast paced but the instructor never lost any of the delegates"
Brian Leek, Data Analyst, May 2022
