Microsoft Azure Data Scientist AssociateStudy guide
The associate certification for designing and implementing ML solutions with Azure Machine Learning (DP-100).
About Microsoft Azure Data Scientist Associate (DP-100)
Microsoft Azure Data Scientist Associate (DP-100) is a Associate-level certification from Microsoft. This page organizes the exam scope into a 3-chapter, 7-section study guide and lets you check your understanding with exam-style practice questions. A good flow is to read the chapters below in order, then test yourself via "Practice questions."
Exam domains (approximate weighting)
- Design and prepare a machine learning solution~23%
- Explore data and train models~38%
- Prepare a model for deployment~24%
- Deploy and retrain a model~15%
Weights are approximate guidance for the live exam. Each domain is covered in detail in the chapters and sections below.
Note: This exam has been retired (successor: AI-300).
Official exam information: https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-100
1Design and Prepare a Machine Learning Solution
- 1.1Azure Machine Learning Workspace and Resources
Understand the core ML platform Azure Machine Learning (Azure ML): the workspace, compute (instances/clusters), datastores/data assets, and dev surfaces (Studio, SDK, CLI). The starting point for DP-100.
- 1.2The ML Process and Responsible AI
Understand the ML lifecycle (data prep → train → evaluate → deploy → monitor), choosing among Automated ML (AutoML), the Designer, and code (SDK/CLI), plus Responsible AI (fairness, explainability).
- 1.3Managing Security, Cost, and Compute
Understand Azure ML operations—authentication/access (Entra/RBAC, managed identity), secrets (Key Vault), networking, compute cost management, and quotas. Operate ML securely and efficiently.
2Explore Data and Train Models
- 2.1Running and Tracking Jobs (MLflow)
Understand training models with Azure ML jobs (command/sweep/pipeline), tracking metrics/parameters/artifacts via MLflow, environments (dependencies), and reuse via components.
- 2.2AutoML and Hyperparameter Tuning
Understand AutoML for automated model search, hyperparameter tuning (sweep jobs)—search space, sampling, early termination—and evaluation metrics/model selection. Find the best model efficiently.
3Deploy and Retrain a Model
- 3.1Registering and Deploying Models (Endpoints)
Understand registering trained models (model registry) and deploying them to inference endpoints—real-time online endpoints and batch endpoints. Put models into production.
- 3.2Monitoring, Retraining, and MLOps
Understand post-deployment monitoring (data drift, performance, inference logs), automating retraining, and MLOps (pipelines, CI/CD, model versioning). Keep models healthy over time.