Databricks GenAI Engineer vs ML Associate
Databricks has two associate-level certifications that people confuse: Generative AI Engineer Associate and Machine Learning Associate. They sit at the same level, cost the same, and test almost entirely different work.
Neither is a prerequisite for the other.
Side by side
| GenAI Engineer Associate | ML Associate | |
|---|---|---|
| Scored questions | 45 | 48 |
| Duration | 90 minutes | 90 minutes |
| Fee | $200 | $200 |
| Validity | 2 years | 2 years |
| Recommended experience | 6+ months with generative AI | 6+ months with ML tasks |
| Centre of gravity | RAG applications and LLM chains | Traditional model development |
| Language | Python | Python (some SQL for data work) |
The sections tell the story
GenAI Engineer — building an LLM application:
| Section | Weight |
|---|---|
| Design Applications | 14% |
| Data Preparation | 14% |
| Application Development | 30% |
| Assembling and Deploying Apps | 22% |
| Governance | 8% |
| Evaluation and Monitoring | 12% |
ML Associate — building a model:
| Section | Weight |
|---|---|
| Databricks Machine Learning | 38% |
| ML Workflows | 19% |
| Model Development | 31% |
| Model Deployment | 12% |
The GenAI exam is organised around an application lifecycle. The ML exam is organised around the platform and the modelling workflow, with Databricks Machine Learning — AutoML, Unity Catalog, MLflow — as its largest single section at 38%.
What each one actually tests
GenAI Engineer assumes you are building something that uses a language model: chunking and indexing documents, constructing a chain, retrieving context, deploying the application, governing what it can do, and evaluating whether its answers are any good. Retrieval-augmented generation is the recurring pattern.
ML Associate assumes you are building a model from data: exploration, feature engineering, training, tuning, tracking experiments with MLflow, registering a model and deploying it. No language models required.
Which should you take?
GenAI Engineer if your work involves LLM applications, RAG over company content, or chat interfaces on Databricks. This is the newer, more in-demand skill set.
ML Associate if you build predictive models — forecasting, classification, recommendation — and your day involves MLflow and AutoML more than prompts and retrieval.
Both, if you work across the platform. They overlap on MLflow, Unity Catalog and deployment concepts, so the second is cheaper in effort than the first. Take whichever matches your current work first, because the hands-on familiarity carries the exam.
The overlap, honestly
Real but modest. Both expect Python, both touch MLflow and Unity Catalog, both cover deployment and monitoring in some form, and both assume you have used Databricks rather than read about it.
The shared 20–30% is mostly platform knowledge. The rest is genuinely different work — building a chain that retrieves documents is not the same skill as engineering features and tuning a gradient-boosted tree.
If you are choosing a first Databricks certification
Pick by what you do this month, not by which sounds more current. Both are 90-minute, $200, two-year credentials with a 6+ month experience expectation, so the deciding factor is which set of tasks you can already perform. Hands-on familiarity is what passes these exams.