Demystifying Notebooks in Fabric

Microsoft Fabric introduces Notebooks as a code-first environment for data engineers and analysts. Many teams rely on visual tools like Dataflows or Power Query for transformations, but these reach limits with massive datasets, unstructured data, or complex logic. Notebooks address these gaps by supporting Python, T-SQL, and Spark for scalable processing.

Jessica Jolly, an experienced trainer at the Academy, leads this session. She clarifies what Notebooks are, when to use them, and how they integrate with Lakehouses. The discussion features real-world demos and input from Jon Manderville and community regulars, covering practical scenarios without overwhelming beginners.

What You Will Learn

• Core concepts of Notebooks in Fabric and key differences from Dataflows or Power Query
• Specific scenarios where Notebooks outperform visual ETL tools
• Why investing time in Python or T-SQL pays off for Fabric workflows
• Techniques for processing structured and unstructured data using Lakehouses
• How Spark enhances performance for large-scale data operations
• Best practices for documenting code, running tests, and sharing notebooks
• Hands-on demos including Excel file uploads, markdown cells for notes, and debugging steps

Key Takeaways

1. Choose Notebooks when visual tools like Power Query hit scalability limits. Jessica explains that for datasets exceeding gigabytes or requiring custom logic, code-based Notebooks with Spark deliver faster results and better control.
2. Pair Notebooks with Lakehouses for flexible data handling. Use them to ingest Excel files directly, blend structured tables with unstructured files, and query everything via Python or T-SQL without data duplication.
3. Document inline with markdown cells. This practice turns notebooks into self-contained reports, making it easy for teams to review transformations, understand assumptions, and collaborate on iterations.
4. Debug iteratively using cell-by-cell execution. Instead of running entire scripts, execute and test individual cells to isolate issues quickly, a technique Jessica demonstrates with common errors like data type mismatches.

About the Speaker

Jessica Jolly is a highly regarded trainer at the Academy, specializing in Microsoft Fabric and Power BI. She delivers approachable sessions that bridge GUI tools and code-based analytics, helping professionals build confidence in new Fabric features. Her style emphasizes practical demos and community interaction.

Who Should Watch This

This session suits Power BI users dipping into Microsoft Fabric who wonder about Notebooks. You will gain clarity if you handle growing datasets in visual tools but face performance bottlenecks or need custom transformations.

Fabric-curious analysts and data professionals will benefit, especially those debating a shift from no-code interfaces to Python or T-SQL. The demos provide a low-pressure entry point to code-driven workflows.

Skip if you already author Spark jobs or manage production Lakehouses daily. However, Jessica's tips on documentation and sharing may refine your existing processes for better team collaboration.