Skip to content
Cheat sheets
Professional · Professional Machine Learning EngineerPrep available

Google Cloud Certified Professional Machine Learning Engineer (Professional Machine Learning Engineer) practice exam and study path

A free, timed 60-question Professional Machine Learning Engineer practice exam in 120 minutes, scored by domain, with every option explained after you submit. Also 116 practice questions you can check one at a time, a page for every task in the exam guide, and a study path. No sign-up.

Start the free practice examGet the exam packPractice question by question →
Professional ML Engineer exam pack:

$11.99for 12 months

The Google Cloud machine learning exam: low-code models with BigQuery ML, AutoML and Gemini, preparing data and prototyping in notebooks, training and tuning at scale on Gemini Enterprise Agent Platform, serving and scaling models, automating pipelines and retraining, and monitoring AI solutions for risk, drift and quality.

The exam

Exam code
Professional Machine Learning Engineer
Questions
Not yet published
Time
120 minutes
Passing score
Not published
Exam fee
$200 (USD)
Question types
multiple choice, multiple response

50-60 multiple-choice questions in 2 hours, $200 plus tax where applicable, online-proctored or at a testing center, in English and Japanese. No prerequisites; 3+ years of industry experience including 1 or more years designing and managing solutions using Google Cloud recommended; valid for 2 years. Holders can renew with designated courses and skill badges in Google Skills (adds 1 year) or with the standard exam from 60 days before the inactive date (adds 2 years); no shorter renewal exam is offered. The exam was updated for the move from Vertex AI to Gemini Enterprise Agent Platform; the exam guide lists the product names used on the exam. Passing score: Not published. Our forms hold 60 questions, the top of the published range, split by the guide's approximate section weights.

From Google Cloud's certification page and exam guide, checked Oct 31, 2026.

Domains and weights

  • Domain 1: Architecting low-code AI solutions

    13%

  • Domain 2: Collaborating within and across teams to manage data and models

    16%

  • Domain 3: Scaling prototypes into ML models

    20%

  • Domain 4: Serving and scaling models

    20%

  • Domain 5: Automating and orchestrating ML pipelines

    18%

  • Domain 6: Monitoring AI solutions

    13%

On our roadmap this comes after Associate Cloud Engineer. See every certification.

Practice

116 original practice questions, every option explained with the Google Cloud documentation page that proves it. Free, no sign-up; progress stays on this device.

Your results on this device →

More practice exams

The lessons, the sample questions on every task page and the first full practice exam are free and stay free. A pack adds the further timed Professional Machine Learning Engineer practice exams.

Professional ML Engineer exam pack

$11.99for 12 months

  • Every paid Professional Machine Learning Engineer practice exam, timed, with every option explained
  • The next version of this exam included when it changes within your 12 months
  • 12 months of access. One payment, nothing renews by itself.
  • 14-day refund, no conditions
  • 14-day refund, no conditions: ask within 14 days of buying and you get the whole payment back.
  • No pass guarantee. These are practice exams written by us from the exam guide and the Google Cloud documentation; a score here does not predict your result on the real exam.

Study path

One module per exam domain. Lessons that already exist on BytePatterns are linked; the rest are being written.

  1. Module 1: Architecting low-code AI solutions

    1. 1Choosing a BigQuery ML or AutoML model: classification, regression, ARIMA_PLUS forecasting and k-meansTask 1.1

      Partly covered by: What Is Machine Learning, Training vs Inference

    2. 2Feature engineering, selection and prediction in BigQuery ML SQLTask 1.1

      Lesson coming

    3. 3Fine-tuning Gemini from BigQueryTask 1.1

      Partly covered by: Fine-Tuning vs Prompting

    4. 4Selecting a model in Model Garden: Gemini, open models and models as a serviceTask 1.2

      Partly covered by: What Is an LLM, Model Routing and Fallbacks

    5. 5Document AI, Vision and Translation APIs for industry use casesTask 1.2

      Lesson coming

    6. 6Tuning Gemini, Imagen and Veo for a use caseTask 1.2

      Partly covered by: Fine-Tuning vs Prompting, Adapters and LoRA

    7. 7Gemini cost, latency and availability: context caching, batch inference, Provisioned Throughput and the global endpointTask 1.2

      Partly covered by: Context Windows, Model Routing and Fallbacks

  2. Module 2: Collaborating within and across teams to manage data and models

    1. 8Organizing tabular, text and image data; BigQuery, Dataflow, Spark or pandas for preprocessingTask 2.1

      Partly covered by: Data Readiness for AI

    2. 9Feature groups, feature views and online stores in Agent Platform Feature StoreTask 2.1

      Lesson coming

    3. 10PII in training data: Sensitive Data Protection, masking and de-identificationTask 2.1

      Partly covered by: Finding and Masking PII

    4. 11Securing and sharing Workbench instances and Colab Enterprise runtimesTask 2.2

      Lesson coming

    5. 12Prototyping with foundation and open models from a notebookTask 2.2

      Partly covered by: Temperature and Sampling, What Is an LLM

    6. 13Experiments, TensorBoard, ML Metadata and lineageTask 2.3

      Lesson coming

    7. 14Evaluating predictive and generative models: metrics, rubrics and LLM-as-a-judgeTask 2.3

      Partly covered by: Evaluating LLMs, LLM as a Judge

  3. Module 3: Scaling prototypes into ML models

    1. 15Choosing the model type and the product: ARIMA, DNN, LLM; AutoML, BigQuery ML or a pipelineTask 3.1

      Partly covered by: What Is Machine Learning, Transformers: Big Picture

    2. 16Interpretable models when explanations are a requirementTask 3.1

      Lesson coming

    3. 17Organizing and ingesting training data in Cloud Storage and BigQueryTask 3.2

      Partly covered by: S3: Consistency, Classes, Lifecycle

    4. 18Serverless training, prebuilt and custom containers, Kubeflow on GKE and Tabular WorkflowsTask 3.2

      Partly covered by: Containers vs VMs, Images, Layers & Multi-Stage Builds

    5. 19Troubleshooting failed training and hyperparameter tuningTask 3.2

      Lesson coming

    6. 20When to fine-tune a foundation model, and howTask 3.2

      Partly covered by: Fine-Tuning vs Prompting, Adapters and LoRA

    7. 21CPUs, GPUs and TPUs; data and model parallelismTask 3.3

      Lesson coming

  4. Module 4: Serving and scaling models

    1. 22Batch and online inference on Agent Platform, Cloud Run and GKETask 4.1

      Partly covered by: Training vs Inference, Pods, ReplicaSets & Deployments

    2. 23Prebuilt and custom serving containersTask 4.1

      Partly covered by: Images, Layers & Multi-Stage Builds

    3. 24Model Registry versions and aliases; A/B tests and canary rolloutsTask 4.1

      Partly covered by: Scheduling & Rolling Updates

    4. 25Inference preprocessing and postprocessingTask 4.1

      Lesson coming

    5. 26Serving features online from Feature StoreTask 4.2

      Lesson coming

    6. 27Public, dedicated and private endpointsTask 4.2

      Lesson coming

    7. 28Serving hardware and autoscaling for throughputTask 4.2

      Partly covered by: Requests, Limits & HPA

    8. 29Making a model cheaper and faster to serve: quantization, speculative decoding and the KV cacheTask 4.2

      Partly covered by: Quantization, Speculative Decoding, The KV Cache

  5. Module 5: Automating and orchestrating ML pipelines

    1. 30Validating data and models inside a pipelineTask 5.1

      Lesson coming

    2. 31Agent Platform Pipelines, Managed Service for Apache Airflow and RayTask 5.1

      Partly covered by: Design a Job Scheduler

    3. 32The same preprocessing for training and servingTask 5.1

      Lesson coming

    4. 33Retraining policies: schedules, triggers and performance thresholdsTask 5.2

      Partly covered by: From Pilot to Production

    5. 34CI/CD/CT for models with Cloud BuildTask 5.2

      Lesson coming

  6. Module 6: Monitoring AI solutions

    1. 35Securing AI systems: exfiltration, malicious prompts, sensitive data, safety filters and Model ArmorTask 6.1

      Partly covered by: Guardrails, Finding and Masking PII

    2. 36Responsible AI: bias and explainabilityTask 6.1

      Lesson coming

    3. 37Model Monitoring: training-serving skew, data drift, concept drift and attribution driftTask 6.2

      Partly covered by: Observability Basics

    4. 38Monitoring, testing and evaluating generative AI in productionTask 6.2

      Partly covered by: Evaluating LLMs, LLM as a Judge

Every task in the exam guide

The exam guide splits each domain into task statements. Each one has a page with a short summary, the lessons that teach it and open sample questions.

Domain 1: Architecting low-code AI solutions

Domain 2: Collaborating within and across teams to manage data and models

Domain 3: Scaling prototypes into ML models

Domain 4: Serving and scaling models

Domain 5: Automating and orchestrating ML pipelines

Domain 6: Monitoring AI solutions

How these questions are made

  • Written by us from the exam guide's task statements and the Google Cloud documentation — never from real exam content.
  • Every option carries its own explanation and a link to the Google Cloud page that proves it.
  • A question enters a timed exam only after an independent check; until then it is a practice question.
  • Results are a plain percentage, per domain too — not an imitation of Google Cloud's scaled score.

Questions about the Professional Machine Learning Engineer practice exam

Is the Professional Machine Learning Engineer practice exam free?
Yes. The full practice exam (60 questions, 120 minutes) is free with no sign-up, and so are the 116 practice questions, the task pages and the study path. Your attempts are saved in this browser, on this device.
How many questions does it have, and is it timed?
60 questions in 120 minutes: the real Professional Machine Learning Engineer pace, scaled to this form. When the time runs out, the exam is submitted as it stands.
Are the answers explained?
After you submit, not during: like the real exam, the timed exam gives no feedback until the end. Then you get your score overall and per exam domain, and every question with each option explained and a link to the Google Cloud documentation page behind it. In practice mode an answer is explained as soon as you check it.
What does the Professional ML Engineer exam pack add?
3 more timed practice exams (Practice exam 1: 60 questions, 120 minutes; Practice exam 2: 60 questions, 120 minutes; Practice exam 3: 60 questions, 120 minutes), each scored and explained the same way, and the next version of this exam if it changes within your 12 months. 12 months of access. One payment, nothing renews by itself. Everything above stays free. See the pack.
Can I get a refund?
Yes, with no conditions: ask within 14 days of buying and you get the whole payment back. The details are in the terms.
Is this an official Google Cloud practice exam?
No. BytePatterns is not affiliated with, endorsed or sponsored by Google Cloud. Every question is written by us from the public exam guide and the Google Cloud documentation, never from real exam content, and a score here does not predict your result on the real exam. Google Cloud lists its own preparation resources on its certification page.