Microsoft Fabric Essentials is a hands-on course in building a data platform on Microsoft Fabric, from OneLake and lakehouses to pipelines, Dataflow Gen2, notebooks, a medallion architecture, warehouses and Direct Lake reports in Power BI, with security and governance through Purview. It suits data analysts, BI teams and data engineers who are new to Fabric.
Data teams in many organisations still work across scattered tools: raw data lives in one place, ETL jobs in another, the data warehouse in a third system, and Power BI reports pull the same data again and again. The result is conflicting copies of data, systems that are hard to maintain and duplicated costs. Microsoft Fabric brings this work together on one platform, with OneLake as the organisation's central data store, so Data Factory, lakehouses, warehouses, Real-Time Intelligence and Power BI all read the same data. Getting value from Fabric, however, means knowing which item to use for which job and structuring data properly from the start.
This course takes data analysts, BI teams and data engineers who are new to Fabric through building an end-to-end data solution. Learners start with OneLake, workspaces and capacities, build a lakehouse, bring data in with pipelines, Copy job and Dataflow Gen2, transform it with notebooks and basic PySpark, organise it in a medallion architecture, and use the warehouse and SQL analytics endpoint. They then build a Direct Lake semantic model and Power BI reports, get an overview of Real-Time Intelligence and Copilot in Fabric, and look after security and governance with Microsoft Purview. The course closes with a capstone that builds a data solution from ingestion to reporting. (3 days, 6 hours per day, 18 hours in total, Intermediate level.)
Course Details
This course is a complete rewrite that replaces the previous Microsoft Fabric course, updated for Microsoft Fabric in 2026. It runs for 3 days, 6 hours per day (18 hours in total, 09:00-16:00), as lectures with labs built on a single sample company dataset throughout. Intermediate level. Learners do the labs on a Fabric trial or their organisation's capacity. It covers OneLake, workspaces and capacities, lakehouses and shortcuts, pipelines, Copy job and Dataflow Gen2, notebooks and basic PySpark, the medallion architecture, the warehouse and SQL analytics endpoint, Direct Lake semantic models and Power BI reports, an overview of Real-Time Intelligence, Copilot in Fabric, and security and governance with Microsoft Purview, and closes with an end-to-end capstone. It is not an official exam-preparation course.
Section 1: Getting to Know Microsoft Fabric and OneLake
The problems with fragmented data systems and the idea of a unified data platform
Fabric workloads: Data Factory, Data Engineering, Warehouse, Real-Time Intelligence and Power BI
OneLake as the central data store, and the Delta Parquet format
Fabric items and how they relate to the Power BI you already use
Lab: explore the Fabric interface and the OneLake catalog
Section 2: Workspaces, Capacities and Organising Work
How tenants, capacities and workspaces differ
Fabric F SKU capacities, trials and a first look at resource usage
Workspace roles: Admin, Member, Contributor and Viewer
Ways to split workspaces by team, by domain or by Dev and Prod environment
Lab: create a project workspace and assign user roles
Section 3: Lab: Lakehouses and Shortcuts
The structure of a lakehouse: Files and Tables
Upload CSV files and turn them into Delta tables
Shortcuts to data in OneLake or external storage without copying it
An overview of mirroring to bring database data into OneLake
Lab: build the sample company's lakehouse and load the first dataset
Section 4: Lab: Ingesting Data with Pipelines and Copy Job
Which jobs suit pipelines, Copy job and Dataflow Gen2
The Copy activity and connecting to data sources through connections
Copy job for full and incremental copies of data
Parameters and control flow activities at a basic level
Lab: a pipeline that loads sales data from a database into the lakehouse
Section 5: Lab: Dataflow Gen2
Power Query Online in Dataflow Gen2 for those who know Power BI
Clean data, merge tables and change data types
Set the data destination to a lakehouse or warehouse
Choosing between Dataflow Gen2 and notebooks by data size and complexity
Lab: transform customer and product data and write it to the lakehouse
Section 6: Lab: Notebooks and PySpark Basics
Notebooks in Fabric and the Spark session
Read and write Delta tables with PySpark DataFrames
Select, filter, join and group by with PySpark and Spark SQL
Use Copilot in notebooks to help write and explain code
Lab: clean transaction data and save it as a new table
Section 7: The Medallion Architecture
What the bronze, silver and gold layers are for
Design lakehouses and workspaces to support each layer
Delta Lake: MERGE, time travel and table maintenance with OPTIMIZE and VACUUM
Design gold tables as a star schema for reporting
Lab: move data from bronze to silver and gold
Section 8: Lab: The Warehouse and T-SQL
How a Fabric warehouse differs from a lakehouse, and when to use it
Create tables, load data and write T-SQL in the warehouse
Views and stored procedures to prepare data for reports
Query across lakehouse and warehouse data
Lab: build fact and dimension tables in the warehouse
Section 9: The SQL Analytics Endpoint
How a lakehouse's SQL analytics endpoint works
Query lakehouse data with read-only T-SQL
Create views and connect from SQL Server Management Studio or VS Code
Lab: write sales analysis queries on the gold tables
Section 10: Lab: Orchestration and Scheduling
Combine Dataflow Gen2, notebooks and stored procedures in one pipeline
Schedule runs and handle what happens when an activity fails
Send run notifications by email or Teams
Track runs in the Monitoring hub
Lab: a pipeline that runs the whole process from bronze to gold automatically
Section 11: Lab: Direct Lake Semantic Models
How Import, DirectQuery and Direct Lake differ
Build a Direct Lake semantic model from the gold tables
Relationships and basic DAX measures
Direct Lake on OneLake versus Direct Lake on SQL, and fallback to DirectQuery
Lab: build a sales semantic model with the key measures
Section 12: Lab: Power BI Reports on Fabric
Build reports in the Power BI service and Power BI Desktop
Use Copilot in Power BI to help create report pages and summarise data
Share reports through a workspace app and set viewer permissions
Lab: a sales dashboard that stays current with data in OneLake
Section 13: Real-Time Intelligence and Copilot in Fabric
An overview of the Real-Time hub, eventstreams and eventhouses
Query real-time data with basic KQL
Activator for alerts when data meets a condition
Copilot and data agents in Fabric for asking questions of data in natural language
Lab: monitor real-time data and set up an alert
Section 14: Security, Governance and Microsoft Purview
Workspace, item and OneLake security permissions
Row-level and column-level security in the warehouse and semantic models
Sensitivity labels, lineage and endorsement of data
Connect Fabric to Microsoft Purview to govern data across the organisation
An overview of Git integration and deployment pipelines to separate Dev and Prod
Section 15: Workshop: End-to-End Capstone
Brief: build a sales data solution for the sample company from start to finish
Ingest data, transform it through the medallion layers and schedule it with a pipeline
Build a Direct Lake semantic model and a report with the right permissions
Present the work, review it together and wrap up with a checklist for adopting Fabric
No public rounds are open right now. Join the waiting list and we will contact you first when the next round opens, or ask us on LINE. Or call 02-570-8449 or 088-807-9770
A medallion architecture organises data into layers by quality. Bronze holds raw data as it arrives from the source, silver holds data that has been cleaned, typed and joined, and gold holds summarised tables ready for reporting, usually designed as a star schema of facts and dimensions. Separating the layers lets you trace problems back and reprocess data without pulling it from the source again.
A lakehouse stores both files and Delta tables in OneLake and is transformed with Spark or notebooks, so it suits varied data and data engineering work; its SQL analytics endpoint lets you query it with read-only T-SQL. A warehouse supports full read and write T-SQL, views and stored procedures, so it suits SQL-focused teams. Both keep their data in OneLake.
Import copies data into the semantic model, which is fast but needs scheduled refreshes. DirectQuery sends queries to the source each time a report is viewed, so data is current but usually slower. Direct Lake reads Delta tables in OneLake straight into memory, giving speed close to Import without copying the data, although in some cases it can fall back to DirectQuery.
It suits data analysts and Power BI developers moving into data platforms on Fabric, data engineers new to Fabric or coming from Azure Data Factory or Synapse, and BI teams that want ETL, warehousing and reporting in one place. You should write basic SQL and have built reports in Power BI Desktop; Python experience is not required.
You will be able to build lakehouses and load data with pipelines, Dataflow Gen2 and notebooks, organise it in medallion layers on a schedule, create Direct Lake semantic models and Power BI reports on OneLake data, and set up security and governance with Microsoft Purview. Participants who complete the training receive a Certificate of Completion with its own number, which can be verified online.
Yes. It can be run in-house for your team, with examples adapted to your own work, such as practising on your organisation's capacity with the kind of data your team looks after. You can request a quotation from this page, or contact us on LINE @itgenius or by phone on 02-570-8449.
Many organisations keep their data spread across Excel workbooks, CSV files and several databases. Every month end, staff copy, paste and reformat the same reports, figures from different departments do not match, and managers do not see the big picture in…
A good dashboard does more than show what sales were. It should help answer why the numbers changed, which customers are about to leave and which points are unusual enough to need attention now. Power BI has built-in AI visuals that answer these questions…
Many organizations already own BI tools such as Power BI or Tableau, yet still struggle with data preparation: pulling data from several systems by hand, cleaning it in Excel, and constantly finding that numbers do not match between reports. This course fills…
