Microsoft Generative AI Data Analysis Course

Lesson 1 • Setting Up Your Analysis Environment

Guides environment configuration for Python, VS Code, and Copilot. A working setup is required before any hands-on coding exercises begin.

Lesson 2 • Microsoft AI Tools Landscape

Surveys Microsoft's AI-integrated products relevant to data work. Students identify which tools apply to their specific analysis scenarios.

Lesson 3 • Data Analysis in the Modern Workplace

Defines data analysis roles, outputs, and business value. Anchors the course by showing where coding and AI accelerate analyst productivity.

Lesson 1 • Working with Files and Data Formats

Demonstrates reading and writing CSV, JSON, and Excel files in Python. Directly enables ingestion of real analyst data in later chapters.

Lesson 2 • Python Syntax and Data Types

Covers variables, operators, and core data types used in analysis scripts. Provides the syntax foundation needed for all subsequent coding chapters.

Lesson 3 • Control Flow and Functions

Teaches conditionals, loops, and reusable functions for automating repetitive analysis steps. Students refactor manual tasks into callable functions.

Lesson 4 • Using Copilot to Accelerate Python Coding

Applies GitHub Copilot to speed up Python script writing with inline suggestions. Students learn to evaluate, accept, and refine AI completions critically.

Lesson 1 • Filtering, Sorting, and Transforming Data

Covers boolean filtering, multi-column sorting, and column transformations. These operations form the core of exploratory data analysis workflows.

Lesson 2 • Data Cleaning Techniques

Addresses missing values, duplicates, and inconsistent formats that impair analysis. Students apply Copilot prompts to generate and validate cleaning code.

Lesson 3 • Pandas DataFrames and Series

Introduces the DataFrame and Series structures central to tabular data work. Establishes indexing and selection patterns used throughout the chapter.

Lesson 4 • Grouping and Aggregation

Teaches groupby operations and pivot tables for summarising business data. Students generate summary reports that feed directly into visualisation chapters.

Lesson 1 • Prompting for Code Generation

Focuses on generating Python and SQL code through precise natural-language prompts. Students compare outputs across prompt variations to understand sensitivity.

Lesson 2 • Chain-of-Thought and Multi-Step Prompts

Introduces chained prompts that break complex analysis into sequential AI-assisted steps. Students build prompt chains for end-to-end data workflows.

Lesson 3 • Prompt Templates and Reusable Libraries

Guides creation of parameterised prompt templates stored for team reuse. Establishes a personal prompt library as a professional productivity asset.

Lesson 4 • Prompting for Data Interpretation

Uses AI to summarise statistics, explain anomalies, and draft narrative insights. Connects AI language output to the storytelling phase of analysis.

Lesson 1 • Copilot-Assisted Automation Script Writing

Uses Copilot to draft, refine, and document automation scripts end to end. Students evaluate AI-generated automation code for correctness and security.

Lesson 2 • Microsoft Graph API for Office Data

Introduces the Microsoft Graph API for reading SharePoint, OneDrive, and Teams data. Students authenticate and retrieve organisational data programmatically.

Lesson 3 • Scheduling and triggering automation scripts

Teaches task scheduling with Windows Task Scheduler and Azure Functions triggers. Students deploy scripts that run automatically on a defined cadence.

Lesson 4 • Excel Automation with openpyxl and xlwings

Covers programmatic workbook creation, formatting, and formula injection. Replaces manual Excel work with reproducible, scheduled Python scripts.

Lesson 1 • AI-Assisted SQL Query Writing

Uses Copilot and natural-language prompts to generate and explain SQL queries. Students validate AI-generated SQL against expected results before deployment.

Lesson 2 • Connecting Python to Databases

Demonstrates SQLAlchemy and pyodbc connections to SQL Server and Azure SQL. Students run queries from Python and load results directly into DataFrames.

Lesson 3 • Building Automated Data Pipelines

Chains SQL extraction, Python transformation, and load steps into scheduled pipelines. Students deploy a working extract-transform-load pipeline by section end.

Lesson 4 • Query Optimisation and Performance

Covers indexing, execution plans, and query rewriting for performance gains. AI tools assist in identifying slow query patterns and suggesting improvements.

Lesson 1 • Interactive Dashboards with Plotly and Streamlit

Builds interactive web dashboards deployable inside a browser or Teams tab. Extends static charts into explorable tools for non-technical stakeholders.

Lesson 2 • Building Charts with Matplotlib and Seaborn

Produces bar, line, scatter, and heatmap charts using Python libraries. Students customise and export publication-quality figures from DataFrame data.

Lesson 3 • AI-Generated Narrative Summaries

Uses Copilot and Azure OpenAI to generate written insight summaries from chart data. Students review, edit, and embed AI narratives into stakeholder reports.

Lesson 4 • Visualisation Principles for Analysts

Covers chart selection, colour theory, and clarity principles for business audiences. Prevents common visualisation mistakes before any code is written.

Lesson 1 • Retrieval-Augmented Generation for Data

Combines vector search with AI generation to answer questions over large document sets. Students build a retrieval-augmented generation pipeline on organisational data.

Lesson 2 • Azure OpenAI API for Custom Analysis

Calls Azure OpenAI endpoints from Python to embed generative AI in analysis tools. Students build functions that send data context and retrieve structured AI responses.

Lesson 3 • AI Governance and Risk Management

Addresses bias, hallucination, data privacy, and auditability in AI-assisted analysis. Students apply a governance checklist before deploying any AI-powered solution.

Lesson 4 • Capstone: End-to-End AI Analysis Project

Students design and deliver a complete AI-assisted analysis project from raw data to stakeholder report. Integrates all core chapter skills into a single assessed deliverable.