DP-100 Azure Data Scientist Associate Training

This course prepares you for the AI-300 certification exam, covering all official exam domains and their approximate weightings:

Domain 1 Design and implement an MLOps infrastructure (17.5%)

• Create and manage a Machine Learning workspace, datastores, and compute targets
• Configure identity and access management (IAM) for workspaces
• Create and manage data assets, environments, and components; share assets across workspaces via registries
• Deploy Machine Learning workspaces and resources by using Bicep and Azure CLI
• Automate resource provisioning with GitHub Actions workflows and restrict network access
• Manage source control for machine learning projects using Git

Domain 2 Implement machine learning model lifecycle and operations (27.5%)

• Configure experiment tracking with MLflow; use automated ML and notebooks for training/exploration
• Automate hyperparameter tuning and run model training scripts as jobs
• Implement training pipelines and manage distributed training for large/deep learning models
• Register MLflow models, package feature-retrieval specs, and evaluate models using responsible AI principles
• Deploy models as real-time or batch endpoints with managed inference; implement progressive rollout/safe rollback
• Monitor production models for data drift and performance, and configure retraining/alert triggers

Domain 3 Design and implement a GenAIOps infrastructure (22.5%)

• Create and configure Microsoft Foundry resources and project environments
• Configure identity/access management (managed identities, RBAC) and network security for Foundry
• Deploy infrastructure using Bicep templates and Azure CLI
• Deploy and manage foundation models via serverless API endpoints and managed compute; select appropriate models
• Configure provisioned throughput units for high-volume workloads and manage model versioning/deployment strategies
• Design, version, and manage prompts (variants, comparison) using Git-based source control

Domain 4 Implement generative AI quality assurance and observability (12.5%)

• Create test datasets and data mapping for comprehensive model evaluation
• Implement AI quality metrics, including groundedness, relevance, coherence, and fluency
• Configure risk and safety evaluations for harmful content detection
• Set up automated evaluation workflows using built-in and custom metrics
• Monitor performance metrics (latency, throughput, response times) and track cost/token consumption
• Configure detailed logging, tracing, and debugging for production troubleshooting

Domain 5 Optimize generative AI systems and model performance (12.5%)

• Optimize RAG retrieval performance by tuning similarity thresholds, chunk sizes, and retrieval strategies
• Select and fine-tune embedding models for domain-specific use cases
• Implement and optimize hybrid search combining semantic and keyword-based retrieval
• Evaluate/improve RAG system performance using relevance metrics and A/B testing
• Design and implement advanced fine-tuning methods, including synthetic data creation
• Manage a fine-tuned model from development through production deployment