Machine Learning at Scale
Databricks
Last updated June 30, 2026
Machine Learning at Scale is a free, two-hour course on the problem that appears when datasets and workloads grow too large for one machine to handle. It centers on Apache Spark — the engine for distributed data processing — and how its architecture applies to large-scale machine learning, so models can be trained across many machines at once. It assumes Python, basic machine learning, and some familiarity with Spark, and the work is done on the Databricks platform. It suits a practitioner whose models have outgrown a single machine.
What you'll learn
- Why scale changes the machine learning problem once data outgrows one machine
- Apache Spark's architecture for distributed data processing
- Applying Spark to large-scale machine learning workloads
- Training models across many machines rather than one
- Carrying it out on the Databricks platform
Frequently asked questions about Machine Learning at Scale
Who is Machine Learning at Scale for?
Practitioners with Python, basic machine learning, and some Apache Spark familiarity who need to run machine learning workloads at large scale.
Is Machine Learning at Scale free?
Yes — Machine Learning at Scale is completely free to take.
What are the prerequisites for Machine Learning at Scale?
Python; basic machine learning; familiarity with Apache Spark; familiarity with Databricks. A free Databricks Academy account is required to start.
Do you need to code for Machine Learning at Scale?
Yes — Machine Learning at Scale involves hands-on coding.
Why we suggest this course
Tackles a specific advanced problem — training models when the data won't fit on one machine — by grounding it in Apache Spark's distributed architecture. One thing to know: it assumes some Spark familiarity going in and is taught on Databricks with a free Academy account, so both the prerequisite and the platform matter.