Databricks ML in Action

Overview

This course offers practical skills to build and deploy machine learning solutions on the Databricks platform, covering the entire ML lifecycle from data ingestion to model deployment. You’ll gain hands-on experience with key tools such as MLflow, Vector Search, and AutoML, while mastering the Databricks Lakehouse architecture. This course will equip you with real-world skills to tackle data science challenges using Databricks' state-of-the-art technologies. The course guides learners through hands-on projects that include tasks like streaming, forecasting, image classification, and retrieval-augmented generation. Whether you’re building machine learning models or deploying them at scale, this course will enhance your proficiency in leveraging Databricks’ robust tools for real-world data science problems. By integrating theory and real-world applications, you’ll learn from practical examples and code projects designed to accelerate your learning process. Unlike traditional courses, this course emphasizes the application of Databricks tools in real business environments, preparing you for both theoretical and hands-on challenges. This course is ideal for data scientists, machine learning engineers, and technical managers with a foundational knowledge of data analysis and machine learning. If you're already familiar with basic data science concepts and cloud environments, this course will elevate your skills in building and operationalizing machine learning data products using Databricks.

Syllabus

Getting Started and Lakehouse Concepts

In this section, we introduce the Databricks Lakehouse architecture, its components, and advantages for ML development, with practical applications through real-world projects.

Designing Databricks Day One

In this section, we explore planning Databricks platform architecture, defining workspace and metastore configurations, and implementing data preparation and feature creation strategies for efficient data and AI workflows.

Building the Bronze Layer

In this section, we explore the Bronze layer in Databricks, focusing on Auto Loader, Delta Live Tables, and Delta Table optimization for efficient data ingestion and transformation.

Getting to Know Your Data

In this section, we cover Delta Live Tables, Lakehouse Monitoring, and Vector Search for data quality and retrieval.

Feature Engineering on Databricks

In this section, we explore Databricks Feature Engineering in Unity Catalog, streaming features with Spark, and point-in-time and on-demand features for real-time model performance.

Tools for Model Training and Experimenting

In this section, we explore building training sets from feature tables, tracking experiments with MLflow, and integrating external models to enhance predictive workflows.

Productionizing ML on Databricks

In this section, we explore deploying ML models using Databricks MLOps inner and outer loops, asset bundles, and registries for scalable and efficient production integration.

Monitoring, Evaluating, and More

In this section, we explore monitoring model inference data, creating visualizations with Lakeview and SQL dashboards, and deploying ML web apps using Hugging Face and Gradio.