Skip to main content
Claude Partner Network - Certified

Gennoor is Certified in the Claude Partner Network.

Certifications & Projects · part of Learning Studio
‹ All Learning Studio courses

Microsoft, project course, intermediate level

Build Real-Time AI on Microsoft Fabric

Train an AI model, serve it live at checkout, track it on a real-time dashboard and get alerts. One hands-on project.

You sign in once to start.

3
modules
27
video lessons
6
quizzes
3.1 h
of video
0:00 / 0:00
  • 5 lectures
  • 9 lab videos
  • 13 demos
  • 6 quizzes

Each bar is one lesson. Gaps separate the modules.

What you will be able to do

  • Explain Fabric in one sentence and describe the tenant, capacity, workspace and item hierarchy
  • Get access with a work account or a personal email, start the trial and say what it includes and leaves out
  • Find your way around Home, workspaces, the New item gallery, OneLake catalog and the Monitoring hub
  • Choose the right item for the job: lakehouse, warehouse, SQL database or eventhouse, and pipeline, Dataflow Gen2, Copy job or notebook
  • Build a bronze, silver and gold lakehouse with shortcuts, a pipeline and notebooks
  • Track experiments and register a model with MLflow
  • Serve predictions through a real-time endpoint and through batch scoring, and use them in an app, a live dashboard, an alert and a Power BI report
  • Run the solution day to day: lineage, monitoring, model versions and cost hygiene

Curriculum

33 items (lessons and quizzes) across 3 modules. Pick any lesson to open it. You will sign in first if you have not already.

Module 1: Tour of Fabric
6 lessons
Module 2: How Fabric Is Useful
14 lessons
Module 3: Project: Brightcart Delivery Promise
13 lessons

About this course

Learn what every major part of Microsoft Fabric is for, then build one working solution that trains a machine-learning model, serves it live to an app, scores data in batch and shows the predictions to the people who need them.

Brightcart is a fictional online retailer. At checkout it promises a delivery date, and about one order in five arrives late. A late order means a refund and a support ticket. You work on that one problem from start to finish. A tour of the platform and your own trial account comes first. Then each Fabric component solves one small piece, and you join the pieces into the Brightcart Delivery Promise: a model that gives every order a late-risk number between 0 and 1.

What you build

  • Data foundation

    a workspace and a lakehouse, OneLake shortcuts, a pipeline that lands five history files, and bronze, silver and gold tables built with notebooks.

  • Model

    two models compared in an MLflow experiment, and the better one registered as a versioned model. It scores about 0.84 on held-out orders, against about 0.81 for the simple baseline.

  • Serving

    a real-time endpoint called by a small checkout app, and a nightly batch pipeline that re-scores open orders.

  • Consuming

    an eventstream and eventhouse, a live dashboard, an Activator alert, and a Power BI report on a Direct Lake semantic model.

How every module works

  1. Story

    A short chapter sets up the problem

  2. Concept

    Built up on screen as a diagram

  3. Code

    Real code with the current SDK names

  4. Lab

    Every step run on camera

  5. Demo

    The same idea in the portal

Three modules (the tour, the components, the project) hold 25 lessons of about 165 minutes, plus two clearly marked optional segments. The first lesson shows the five-step pattern you will build and where companies use the same pattern on their own data. The tour and the component lessons show each tool in the portal, and the project has seven hands-on labs. Animations and five slides explain what is hard to see. Each module ends with two short quizzes.

Everything runs on the Fabric trial. The optional segments need a paid capacity or an Azure subscription; they show how to pause or delete what they create. All the code is provided: you run it and read it. Whenever something starts using capacity, the lesson says so and switches it off.

What you get

3.1

hours of video

5

lectures

9

lab videos

13

portal demos

6

quizzes

  • Captioned lessons in three modules
  • Seven lab guides and two optional guides, as PDF
  • A course map and cheat sheet, the five slides and ten "In one minute" machine-learning cards, as PDF
  • the synthetic Brightcart data, five notebooks, SQL, KQL and DAX files, build notes and the checkout app

Who it is for, and what comes next

The course is for beginners to Fabric who are comfortable with spreadsheets and basic SQL: analysts, junior data engineers, data science students and IT generalists. It is a project course, not exam preparation. Afterwards you can say what each part of Fabric does and run a small end-to-end machine-learning solution with sensible cost habits.

Exam note: the project practises many of the skills measured in Microsoft Exam DP-600 (lakehouse, pipelines, notebooks, SQL and KQL, Direct Lake, Git and deployment pipelines). It is a hands-on project course, not a complete exam-prep course. A skills map in the course documents shows what is covered.