Free
- Untimed practice across all three domains
- Every question explained against the objective it tests
- Domains by name and official weight
Databricks · Machine Learning Professional
Modelling and ML Ops are 44% each. You don't get to be good at one.
Three domains, and two of them carry the same weight: Model Development and ML Ops are 88% of the exam between them, with Model Deployment making up the rest. A strong modeller who has never automated a retraining job is sitting half an exam. 59 questions in 120 minutes, a year of hands-on work recommended, against a pass mark Databricks doesn't publish.
- Weighted to the current three-domain split, not the four-domain outline most prep material still describes
- ML Ops practised at its real weight — testing, bundles, retraining and drift, not as an afterthought
- Python throughout, with MLflow and the Databricks CLI
Free practice across all three domains. No card required.
Machine Learning ProfessionalML OpsHard
A deployed model's accuracy has fallen over six months. Monitoring shows the input feature distributions are statistically unchanged. What has most likely happened?
Select one.
One choice only. Your earliest pick was dropped.
Where this question sits in the exam
ML Ops is 44% of the Machine Learning Professional exam. Our mocks use the same mix.
0 of 1 selected
An original item written against the Databricks exam guide, last checked 19 September 2026. Not a real exam question.
How it's scored
The pass mark isn't just unpublished. It moves.
Databricks sets passing scores statistically and revises them as questions change. That's their stated reason for not printing one — so the figure you've seen quoted didn't come from them.
A readiness read still needs a line, so we use 80% weighted accuracy. It's ours, not theirs, and it's drawn that way: dashed where a published pass mark would be solid.
We set it high on purpose. Being too strict costs you a week of practice. Being too lenient costs you USD 200.
Machine Learning Professional59 questions · 120 min
Accuracy by domain, weighted as the exam weights them · assumed 80% — ours, not Databricks'
Model Development44%72%
ML Ops44%61%
Model Deployment12%Furthest behind54%
Model Deployment is your lowest score. ML Ops is the bigger drag — a smaller gap, but on 44% of the form instead of 12%.
Recertifying here means the full current exam
The credential lasts two years. Databricks publishes a shorter renewal exam for the Data Engineer Associate only. Recertifying here means sitting whatever version of the full 59-question exam is live then.
On a platform shipping this fast the guide will have moved. Features that did not exist when you certified are on the form, and topics you revised for may be gone.
What the questions look like
Production problems, described the way you'd meet them
Symptoms first, cause second
Items hand you a behaviour — accuracy falling while inputs hold steady, two runs disagreeing on the same data — and ask what it means. The named cause matters more than the fix.
Several options deploy successfully
Rollout and environment questions offer approaches that all work. The requirement that separates them is usually reversibility or reviewability, stated in one clause of the stem.
Two domains, no trade
Model Development and ML Ops are weighted identically, so strength in one cannot cover weakness in the other. Together they decide the exam.
The loop
Practice, sit, review, repeat
Step 1 is free. Steps 2 to 4 need a subscription.
certs.acedify.com/machine-learning-professional
Practice · untimed
HardA deployed model's accuracy has fallen over six months. Monitoring shows the input feature distributions are statistically unchanged. What has most likely happened?
Select one.
Feature drift, which the monitor is not configured to detect
Training/serving skew introduced when the model was deployed
Concept drift: the relationship between the features and the label has changed, while the features have not
Label leakage in the original training set
Stable inputs and falling accuracy is concept drift: the mapping from features to outcome has moved. An input-distribution monitor stays quiet because the features have not changed. Domain: ML Ops.
Click a step to see it.
What you get back
No official line. So we show you ours.
Your readiness is a range, because it's an inference from the questions you've answered rather than a result you've earned. For a certification the outcome is binary, so the question is whether the whole range sits above the line.
The difference here is whose line it is. AWS publishes 720 and we measure you against their number. Databricks publishes nothing, so we measure you against ours — 80% weighted accuracy, set high on purpose, drawn dashed and labelled everywhere it appears.
With two domains tied at 44%, the weighting does less work here than on other exams — which means the raw gap between your modelling and your ML Ops scores is close to the whole story.
When the range stops narrowing and stops rising, more drilling isn't buying you anything, and that's worth knowing before you book.
If the real pass mark is below ours, the worst case is that you sat more practice than you needed. We'd rather make that mistake than the other one.
Machine Learning Professional · 4 mocks completed
Where you stand
Borderline
70–90
assumed 80% — ours, not Databricks'
Your range crosses our assumed line. On today's evidence it's a coin flip, and we'll say so.
Three states
Try the three states. We won't show you a percentage chance of passing — nobody can calculate one honestly without years of real outcomes to calibrate against.
What we don't do
We won't show you a percentage chance of passing. Nobody can calculate one honestly without years of real outcomes behind it.
What it actually tests
Three domains, two of them tied
Model Development and ML Ops are 44% each. Model Deployment is 12%. That shape is unusual and it's worth planning around: most exams let a strong domain carry a weak one, and this one gives you two of equal size and nowhere to hide.
| Domain | Weight | What it covers |
|---|---|---|
| Model Development | 44% · about 26 questions | Joint largest. Advanced training and evaluation, experiment design, and the testing strategies that make a model reproducible rather than merely accurate. |
| ML Ops | 44% · about 26 questions | Joint largest, and weighted identically to Model Development. Testing strategies, environment management with Declarative Automation Bundles, automated retraining, and drift detection with Lakehouse Monitoring. |
| Model Deployment | 12% · about 7 questions | The remainder. Deployment strategies, custom model serving and managing a rollout. |
Domain weights are published on the Databricks certification page. Last checked 19 September 2026.
Model Development plus ML Ops is 88% of the form. If you have a limited week, that is where it goes.
Exam facts
Null fields are omitted. We do not print TBA.
- Credential
- Databricks Certified Machine Learning Professional
- Level
- Professional
- Length
- 120 minutes
- Scored questions
- 59
- Unscored questions
- Present, count not published
- Question types
- Multiple choice
- Pass mark
- Not published
- Domains
- 3
- Fee
- USD 200 + local tax
- Delivery
- Online or test centre, proctored
- Test aids
- None
- Languages
- English
- Prerequisite
- None
- Recommended experience
- One year hands-on with the tasks in the exam guide
- Validity
- 2 years
- Retake wait
- 14 days between attempts, unlimited attempts, full fee each
- Recertification
- Full current exam — no renewal exam for this credential
Pricing
Practice free. Pay when you want to know where you stand.
Pro
- Timed mocks at 59 questions in 120 minutes
- Weighted accuracy across every mock you've sat
- Your readiness range, and how it moves
Everything on every exam, one subscription
One subscription covers the Databricks certifications, AWS, the Microsoft certifications, the Salesforce certifications, the Claude certifications, TOEFL, IELTS and the SAT.
The exam fee is separate and goes to Databricks. We're prep, not registration.
Start practising for freeFAQ
Machine Learning Professional questions
Find out which of the two big domains is behind
Free practice across all three domains, no card. The timed mocks and the readiness range are there when you want them.
Start practising for freeAlso available for the Machine Learning Associate, Generative AI Engineer Associate and the rest of the Databricks track on the same subscription.