Advanced3 days
Azure Machine Learning
Build, train, and deploy ML models at enterprise scale
Overview
Azure Machine Learning is Microsoft's end-to-end ML platform for building, training, and deploying models in production. This training covers the full MLOps lifecycle — from dataset management through to automated retraining pipelines and real-time inference endpoints.
What you'll learn
- Set up and navigate Azure ML workspaces and compute resources
- Manage datasets, experiments, and model versioning in Azure ML
- Train models using Azure ML jobs, pipelines, and AutoML
- Deploy models as real-time and batch inference endpoints
- Implement MLOps practices: CI/CD, monitoring, and retraining triggers
- Integrate Azure ML with Databricks, Synapse, and Azure DevOps
Programme
Day 1 — Workspace, data & training
- Azure ML workspace architecture: computes, datastores, environments
- Data assets: registering and versioning datasets
- Training jobs: submitting scripts, logging metrics, and tracking runs
- AutoML: automated model selection and hyperparameter tuning
- Responsible AI: fairness, explainability, and model cards
- Hands-on: train and compare three models with experiment tracking
Day 2 — Pipelines & deployment
- Azure ML pipelines: reusable, modular ML workflows
- Component-based pipeline design for maintainability
- Real-time inference endpoints: deploy, test, and scale
- Batch inference endpoints: scoring large datasets cost-effectively
- Online endpoint monitoring: data drift and performance degradation
- Hands-on: build a multi-step training pipeline and deploy the model
Day 3 — MLOps & production practices
- MLOps maturity model: from ad hoc scripts to production systems
- CI/CD for ML with Azure DevOps and GitHub Actions
- Model registry: versioning, staging, and promotion workflows
- Automated retraining pipelines triggered by data drift
- Cost management and compute right-sizing for ML workloads
- Hands-on: build a full MLOps pipeline from training to deployment
Who is this for?
- Data scientists moving from notebooks to production ML
- ML engineers building and maintaining ML platforms
- Teams implementing MLOps and responsible AI practices
Prerequisites
- Solid Python and machine learning fundamentals
- Experience with scikit-learn, PyTorch, or TensorFlow
- Basic Azure knowledge helpful
Tools & technologies covered
Azure Machine LearningAutoMLMLflowAzure PipelinesAzure DevOpsPython SDK v2Azure Monitor
Not sure which course fits your team?
Talk to us — we'll match you to the right training path.