DP-100: Designing and Implementing a Data Science Solution on Azure

Course Description

Elevate your data science career by mastering Azure Machine Learning for end-to-end data science solutions. This four-day, instructor-led course empowers data scientists to design, build, deploy, and monitor machine learning models at cloud scale. You’ll gain in-depth experience crafting ML pipelines, leveraging MLflow tracking, deploying models to batch and real-time endpoints, and integrating Responsible AI practices. This course aligns with the Microsoft Certified: Azure Data Scientist Associate certification, reinforcing skills with real-world architecture and production-ready workflows.

Target Audience

Ideal for:

• Data Scientists, ML Engineers, and AI Developers deploying and operating ML solutions using Azure Machine Learning and MLflow

• Professionals preparing for the DP100 Azure Data Scientist Associate certification exam

Prerequisites:

• Proficiency in Python and familiarity with machine learning frameworks such as ScikitLearn, PyTorch, or TensorFlow

• Basic understanding of cloud concepts and Azure services like storage and compute 

Course Outline 

Module 1: Design and Create an Azure Machine Learning Workspace

• Set up and configure an Azure Machine Learning workspace using Studio, Python SDK v2, and Azure CLI

• Organize data, compute targets, environments, and registered models for reproducible workflows 

Module 2: Prepare Data and Build Experiments

• Register data assets and use datasets and datastores to feed ML workflows

• Configure experiment compute targets (compute instances, clusters) and encapsulate dependencies in environments

Module 3: Train, Track, and Optimize Models

• Implement custom code experiments and track metrics with MLflow

• Use Automated ML for rapid model generation and run hyperparameter tuning

• Create training pipelines for modular and scalable execution 

Module 4: Manage and Evaluate ML Models

• Register models from both code and Automated ML results

• Analyze model metrics and use Responsible AI dashboards to assess fairness, interpretability, and bias

Module 5: Deploy to Real-Time and Batch Endpoints

• Deploy registered models as managed online endpoints with scaling, authentication, and versioning

• Configure batch endpoints to process large datasets in a serverless fashion 

Module 6: Build ML Pipelines and Operationalize MLOps

• Design production workflows with pipeline steps that integrate compute, data, and model assets

• Incorporate retraining, CI/CD patterns, and orchestration best practices

Module 7: Monitor Models and Enable Responsible AI

• Monitor deployed models for performance with Azure Monitor, MLflow, and built-in drift detection

• Implement model governance, logging, and automated retraining for production reliability

HandsOn Experience

Expect 40–50% of class time dedicated to practical exercises. You’ll build complete Azure data science solutions—from workspace setup to pipeline orchestration, model deployment, and monitoring—with instructor-guided scenarios.

Skills You’ll Gain

By the end of the course, you’ll be able to:

• Structure and manage Azure Machine Learning workspaces and assets

• Train, track, and optimize models using MLflow, Automated ML, and custom pipelines

• Deploy models as real-time and batch endpoints with version control and security

• Embed Responsible AI capabilities and monitor model drift and performance

• Operationalize ML solutions with pipelines, retraining strategies, and CI/CD integration