AI-300 cheat sheet

Exam at a glance

Domain weightings

How much of the exam each domain covers. Spend your study time in proportion — the heavier the domain, the more questions you'll see.

Key terms

MLOps

MLOps (machine learning operations) is the practice of automating and operationalizing the ML lifecycle — training, deployment, monitoring, and retraining. AI-300 centers on building MLOps infrastructure on Azure Machine Learning.

GenAIOps

GenAIOps (generative AI operations) is the operational discipline for deploying, evaluating, monitoring, and optimizing generative AI applications and agents. AI-300 pairs it with MLOps under the umbrella of AI operations.

Azure Machine Learning

Azure Machine Learning is the Azure service for training, deploying, and managing traditional ML models with workspaces, compute, and pipelines. AI-300 uses it as the backbone of MLOps infrastructure.

MLflow

MLflow is an open-source framework for experiment tracking, model packaging, and model registry that Azure Machine Learning uses natively. AI-300 requires tracking experiments and registering MLflow models.

Model registry

A model registry is a versioned catalog of trained models and their metadata used to manage promotion and deployment. AI-300 covers model registration, versioning, and lifecycle management including archiving.

Data drift

Data drift is the change over time in the distribution of production input data relative to training data, which can degrade model accuracy. AI-300 requires detecting drift and configuring retraining or alert triggers.

Managed endpoint

A managed endpoint is an Azure Machine Learning hosted service that serves a model for real-time or batch inference with managed infrastructure. AI-300 covers deploying models to managed endpoints with rollout and rollback strategies.

Progressive rollout

Progressive rollout is a deployment strategy that shifts traffic to a new model version gradually so it can be validated and safely rolled back. AI-300 covers implementing progressive rollout and safe rollback for endpoints.

Bicep

Bicep is a domain-specific language for declaratively deploying Azure resources as infrastructure as code. AI-300 uses Bicep with Azure CLI to provision Machine Learning and Foundry infrastructure.

Infrastructure as code

Infrastructure as code (IaC) is the practice of provisioning and managing infrastructure through version-controlled definition files rather than manual steps. AI-300 requires implementing IaC for Machine Learning with Bicep and Azure CLI.

GitHub Actions

GitHub Actions is a CI/CD automation platform that runs workflows on repository events. AI-300 uses GitHub Actions to automate resource provisioning and machine learning pipelines.

Provisioned throughput

Provisioned throughput units reserve dedicated model capacity for predictable, high-volume generative AI workloads. AI-300 covers configuring provisioned throughput when deploying foundation models to production.

Groundedness

Groundedness is an AI quality metric that measures how well a generated response is supported by the provided source context. AI-300 requires configuring groundedness alongside relevance, coherence, and fluency metrics.

Evaluation

Evaluation is the systematic measurement of generative AI quality and safety using test datasets and metrics. AI-300 covers building automated evaluation workflows with built-in and custom metrics.

RAG

RAG (retrieval-augmented generation) grounds a model's output in retrieved data to improve relevance and accuracy. AI-300 covers optimizing RAG through chunk sizing, similarity thresholds, embedding choice, and hybrid search.