Data science

Hi data_quantum,

Welcome to the Microsoft Fabric community! For a beginner starting with Data Science, I’d recommend learning Fabric gradually rather than trying to cover all workloads at once.

A practical learning path would be:

Python basics → Pandas/NumPy → Data Exploration → Fabric Lakehouse → Fabric Notebooks → PySpark → Machine Learning.

For hands-on practice, you could start with a small project such as customer churn prediction, sales/demand forecasting, or customer segmentation. A simple workflow would be to load a dataset into a Lakehouse, explore and clean it using a Fabric Notebook, perform feature engineering, train a basic ML model, and evaluate the results.

Microsoft Learn is a good starting point for structured Fabric learning, and the Fabric Data Science experience provides a useful environment for practicing notebooks and ML workflows.

My suggestion would be to build one small end-to-end project while learning. This makes it much easier to understand how the different Fabric components fit together than learning each feature separately.

All the best with your Data Science journey!