DP-100 Data Science on Azure

Skill Goal:
Develop an understanding of the data science lifecycle and how Microsoft Azure supports AI-driven business solutions.

Skills Developed:

• Explain the data science lifecycle and its role in business decision-making

• Recognize how data science integrates with AI-driven solutions

• Navigate the Microsoft Azure portal

• Identify Azure services used for data science and machine learning

Sprint Outcome:
Ability to describe how Azure services support end-to-end data science workflows.

Skill Goal:
Configure and manage an Azure-based environment for data science projects.

Skills Developed:

• Organize Azure resources effectively

• Create and manage Azure Machine Learning workspaces

• Configure compute targets and environments

• Manage datasets and data assets within Azure ML

Sprint Outcome:
Ability to set up and manage a structured Azure Machine Learning environment.

Skill Goal:
Explore and analyze datasets within the Azure ecosystem.

Skills Developed:

• Connect to various data sources including databases and cloud storage

• Perform Exploratory Data Analysis (EDA)

• Identify data quality issues and patterns

• Interpret insights that influence modeling decisions

Sprint Outcome:
Ability to assess data readiness and extract meaningful patterns prior to modeling.

Skill Goal:
Prepare and transform data for machine learning within Azure.

Skills Developed:

• Clean datasets by handling missing values and outliers

• Apply feature selection and basic feature engineering concepts

• Use Azure Data Factory for data ingestion

• Utilize Azure Databricks for collaborative data processing

• Build repeatable data preparation workflows

Sprint Outcome:
Ability to prepare high-quality datasets suitable for machine learning development.

Skill Goal:
Understand machine learning principles and map them to Azure services.

Skills Developed:

• Differentiate regression, classification, and clustering problems

• Follow the machine learning lifecycle

• Identify common algorithms and appropriate use cases

• Align machine learning concepts with Azure ML services

Sprint Outcome:
Ability to select appropriate machine learning approaches for business scenarios using Azure.

Skill Goal:
Develop and train machine learning models using Azure ML tools.

Skills Developed:

• Use Azure Machine Learning Studio for model development

• Create experiments and pipelines

• Split data for training, validation, and testing

• Train models using prepared datasets

• Interpret model outputs and evaluation results

Sprint Outcome:
Ability to develop and train machine learning models within Azure ML.

Skill Goal:
Evaluate and improve machine learning model performance.

Skills Developed:

• Apply regression and classification evaluation metrics

• Compare multiple model performances

• Tune hyperparameters to improve accuracy

• Track experiments and results within Azure ML

Sprint Outcome:
Ability to assess model performance and apply optimization techniques.

Skill Goal:
Deploy machine learning models into production environments.

Skills Developed:

• Understand model deployment concepts

• Deploy models as web services

• Manage endpoints and inference workflows

• Apply scalable deployment practices

• Implement versioning best practices

Sprint Outcome:
Ability to deploy and manage machine learning models as Azure-based services.

Skill Goal:
Manage, monitor, and maintain machine learning solutions in production.

Skills Developed:

• Monitor deployed models

• Plan model retraining workflows

• Ensure scalability, reliability, and security

• Understand governance and MLOps concepts

Sprint Outcome:
Ability to operationalize machine learning solutions using Azure best practices.

Skill Goal:
Integrate prebuilt AI services into data science solutions.

Skills Developed:

• Understand Azure Cognitive Services and use cases

• Apply prebuilt AI services for vision, language, and speech

• Combine cognitive services with custom machine learning solutions

• Evaluate real-world AI integration scenarios

Sprint Outcome:
Ability to incorporate Azure Cognitive Services into AI-powered applications.

Skill Goal:
Design complete, enterprise-ready data science solutions on Azure.

Skills Developed:

• Design end-to-end data science workflows

• Integrate data ingestion, model training, deployment, and consumption

• Apply enterprise architecture best practices

• Prepare for DP-100 exam-style scenarios

Sprint Outcome:
Ability to design scalable, enterprise-grade Azure data science solutions.

Project Goal

Apply Microsoft Azure services to design, build, deploy, and present a complete end-to-end machine learning solution for a structured real-world business scenario.

Skills Demonstrated

• Analyze a defined business problem suitable for machine learning

• Design an Azure-based data science architecture

• Configure Azure Machine Learning workspace and compute resources

• Prepare and transform datasets using Azure tools

• Develop and train machine learning models

• Evaluate model performance using appropriate metrics

• Deploy a trained model as a web service endpoint

• Monitor model performance and discuss retraining strategies

• Apply governance, scalability, and MLOps considerations

• Present technical and business recommendations clearly