Databricks Machine Learning: Build, Tune & Deploy ML Models

About this course

This course contains the use of artificial intelligence.

Build a complete, tracked, tuned, and registered machine learning model in Databricks—starting with a Delta table and ending with a portfolio-ready project.

Most Databricks learners can open a notebook and run a query. Far fewer can confidently turn real Spark data into a machine learning workflow that is reproducible, cost-aware, measurable, and ready for team handoff.

In this practical course, you will build a customer churn prediction project using Databricks Runtime ML, Spark, MLflow, Optuna, AutoML, and feature-management practices. You will learn how to move from raw data to model decision without getting lost in disconnected tools, untracked experiments, or expensive trial-and-error tuning.

Who this course is for

Data analysts who want to move from SQL dashboards to machine learning projects

Data engineers who need to prepare reliable Spark data for ML teams

Junior data scientists who want practical Databricks project experienceML engineers who want a structured Databricks workflow for tracking and tuning models

Professionals preparing a portfolio project for a data, analytics, or ML role

What you will learn

Build a practical Databricks machine learning project from a Delta table

Choose the right Databricks Runtime ML and compute approach for tabular ML

Use Spark DataFrames to profile, clean, and prepare model data

Create leakage-safe numerical, categorical, date, and aggregation features

Build a reusable feature-preparation pipeline

Train and evaluate a baseline classification model

Track parameters, metrics, artifacts, and models with MLflowRun Databricks AutoML as a fast benchmark and inspect its generated notebooks

Tune a model with Optuna and compare 20 or more tracked trials<

Prices, discounts and availability are set by Udemy. We may earn a commission when you purchase through links on this site.