Exam Prep
Google Cloud · PML

Professional Machine Learning / AI Engineer

Validates the skills to build, evaluate, productionize, and optimize AI solutions on Google Cloud — covers ML problem framing, data and model preparation, pipeline orchestration, model training and tuning, serving and monitoring, and responsible AI practices.

Start Preparing

Free study references: Pages from Google Cloud for every objective

Should I Take the PML?

You should work towards PML if you’re aiming for one of these careers:

  • Machine learning engineer — designing, building and operationalizing ML models and AI solutions on Google Cloud, from BigQuery ML and AutoML through custom training and serving.
  • MLOps engineer — automating and orchestrating the pipelines that take a model from prototype to production, and retraining it as it drifts.
  • AI/generative AI engineer — selecting, fine-tuning, evaluating and deploying foundation models such as Gemini, Imagen and Veo, and defending them against malicious prompting.
  • Data scientist moving into production ML — taking a model past the notebook, into Feature Store, a managed pipeline, and a monitored endpoint.

The exam sits alongside Google Cloud's Professional Data Engineer and Professional Cloud Architect certifications, and overlaps both: PDE stops at the data reaching the model, PCE asks which AI product an architecture needs, and this asks how the model itself is built, trained, served, retrained and watched. Holding more than one of these professional certifications builds a broader cloud AI portfolio than any single one does.

Because the current guide is written in the newest Google Cloud product names — the Gemini Enterprise Agent Platform, Model Armor, LLM-as-a-judge — studying for it is also a way to get current on generative AI tooling on Google Cloud, not just predictive ML.

Is the PML Worthwhile?

Here’s what we’d tell a friend:

  • It is Google Cloud's professional-tier credential for the ML engineering role. It sits above the associate level, alongside the Professional Cloud Architect and Professional Data Engineer certifications, and its syllabus runs end to end from low-code model building through serving, pipelines and monitoring.
  • Pay figures for the roles it targets vary sharply by source and seniority — treat any single number with caution. Salary.com's wage data put the US average machine learning engineer salary at $109,921 as of November 2026, while a KORE1 salary guide put 2026 base pay at $128,000-$186,000 with senior engineers at large AI labs clearing $350,000 in total compensation once equity and bonuses are added. Those figures mix entry-level and senior roles, base pay and total compensation, and different company tiers. The certification alone will not set your salary; your role, your employer and your experience will. Source: Machine Learning Engineer Salary, Machine Learning Engineer Salary Guide
  • MLOps hiring is steady rather than booming. An analysis of US MLOps job postings found hiring running at around 70 new postings a week through 2026, described as steady but not spiking, with a median salary of $185,000 across the roles tracked. Source: How Hot Is the MLOps Job Market in 2026?
  • The exam leans on conventional ML that Google's own training does not fully cover. The guide asks for knowledge of ARIMA, DNNs, hyperparameter tuning, data and model parallelism, concept drift and evaluation metrics, material that sits outside what Google's own courses spend most of their time on. Expect to fill that gap from general ML study, not just Google Cloud documentation.
  • Pair it with hands-on pipeline and serving work. The syllabus rewards having actually built a Vertex AI pipeline, deployed an endpoint, or tuned a Gemini application for cost and latency, not just having read about those steps.

How Long Will I Have to Spend Studying for the PML?

The honest answer is that it depends on how much you already know. Google Cloud's own description of the candidate says the engineer builds, evaluates, productionizes, and optimizes AI solutions using Google Cloud capabilities and knowledge of conventional ML approaches, and should be proficient in model architecture, data and ML pipeline creation, MLOps, and metrics interpretation, and familiar with prompt and context engineering, application development, infrastructure management, data engineering, and data governance. If that’s your background, you’ve got a good start.

In our plan, you pick the areas of PML that matter to you — for the exam and for your job — and build a study plan from our domain reviews. Then you take an initial exam, backed by Google Cloud’s own documentation, to see where you need to focus to meet your study plan’s goals (our exam review will show you that).

In our store, you can buy a single exam or, if you intend to stick with us, our standard package — which includes, for example, custom drills for the areas our review showed that you need to strengthen (both domain knowledge and exam-taking skills). Repeat with fresh exams, step up to a challenge exam, take advantage of Mentor, and walk into the real exam with your own two-page cheat sheet. See how it works.

How Hard Is the PML Exam?

Another honest answer: since Google Cloud doesn’t publish pass rates, no one knows (including us). The skills it measures, and how much each area weighs, are listed below.

We can tell you that we update our exams monthly, so they always reflect the current state of both the certification and the documentation that supports it. That means our exams follow the certification’s current topic weights, and our domain reviews and explanations are always up to date.

Start Preparing

We can’t tell you how well you know the content — you need to take one of our real exams for that. But our full exams also tell you how you miss: what question answering habits are tripping you up (and we do it across six kinds of ‘bad habits’). We have over 100 answers written that not only test your knowledge but catch those specific bad habits.

These 10 questions, for example, find which kind of ‘almost right’ answers catch you:

Sample exam coming soon.

Skills at a glance

Architecting low-code AI solutions12.9%
  • Building models in BigQuery ML or Agent Platform AutoML (e.g., classification, regression, forecasting, and clustering) based on the business problem
  • Performing feature engineering or selection using BigQuery ML
  • Generating predictions using BigQuery ML
  • Training models using Agent Platform AutoML
  • Fine-tuning Gemini models using BigQuery
  • Evaluating and selecting the appropriate model for a given task from Gemini Enterprise Agent Platform Model Garden
  • Building applications using industry-specific APIs (e.g., Document AI API, Vision API, and Translate API)
  • Building solutions and tuning models for specific use cases (e.g., Gemini, Imagen, Veo, and models as a service in Model Garden)
  • Optimizing Gemini-based applications for cost, latency, and availability
Collaborating within and across teams to manage data and models15.8%
  • Organizing and exploring different data types (e.g., tabular, text, and images) for efficient experimenting, training, and serving
  • Choosing the right tool for data preprocessing based on scale and complexity (e.g., BigQuery [SQL], Dataflow, Apache Spark, and in-memory Python frameworks)
  • Creating and consolidating features in Gemini Enterprise Agent Platform Feature Store
  • Ensuring data privacy and handling sensitive information (e.g., personally identifiable information [PII])
  • Applying collaboration and security best practices when setting up and running notebook environments
  • Developing models in Agent Platform Workbench or Colab Enterprise notebooks using common frameworks (e.g., PyTorch, sklearn, and JAX)
  • Using a variety of foundational and open-source models in Model Garden to create model prototypes in notebook environments
  • Choosing the appropriate Google Cloud environment for development and experimentation (e.g., Experiments on Gemini Enterprise Agent Platform, Gemini Enterprise Agent Platform Pipelines, and Kubeflow Pipelines) given the framework
  • Evaluating predictive and gen AI solutions (e.g., model evaluation metrics and LLM-as-a-judge)
  • Tracking and comparing model artifacts, versions, and lineage (e.g., Experiments on Agent Platform and Gemini Enterprise Agent Platform ML Metadata)
Scaling prototypes into ML models20.8%
  • Choosing the model type (e.g., ARIMA, DNN, and LLM)
  • Choosing the product (e.g., Agent Platform AutoML, BigQuery ML, and Agent Platform Pipelines)
  • Choosing the deployment strategy
  • Modeling techniques given interpretability requirements
  • Organizing training data (e.g., tabular, text, speech, images, and videos) on Google Cloud (e.g., Cloud Storage and BigQuery)
  • Ingesting structured and unstructured data from various sources into training pipelines
  • Model training using different software development kits (SDKs) (e.g., Agent Platform custom training, Kubeflow on Google Kubernetes Engine [GKE], Agent Platform AutoML, and Tabular Workflows) and organizing training on Google Cloud
  • Troubleshooting ML model training failures
  • Hyperparameter tuning
  • Fine-tuning foundational models from Agent Platform and Model Garden and when tuning should be considered
  • Evaluation of compute and accelerator options (e.g., CPU, GPU, and TPU)
  • Understanding the options for distributed training on GPUs and TPUs using data and model parallelism strategies
Serving and scaling models19.8%
  • Deploying models for batch and online inference using appropriate services (e.g., Agent Platform, Model Garden, Cloud Run, and GKE)
  • Packaging and serving models from different frameworks (e.g., PyTorch and XGBoost) using prebuilt and custom containers
  • Organizing and versioning models in Gemini Enterprise Agent Platform Model Registry
  • Implementing model rollout strategies (e.g., A/B testing and canary deployments) to compare model versions
  • Developing solutions for inference preprocessing and postprocessing
  • Managing and serving features using Agent Platform Feature Store
  • Deploying models to public and private endpoints
  • Choosing appropriate hardware (e.g., CPU, GPU, TPU, and edge)
  • Scaling the serving backend based on the throughput (e.g., Gemini Enterprise Agent Platform Inference and containerized serving)
  • Tuning ML models for training and serving in production
Automating and orchestrating ML pipelines17.8%
  • Validating data and models
  • Building and orchestrating pipelines using managed or unmanaged services and from templates or custom solutions (e.g., Agent Platform Pipelines, Managed Service for Apache Airflow, and Ray on Gemini Enterprise Agent Platform)
  • Ensuring consistent data preprocessing between training and serving
  • Determining an appropriate retraining policy
  • Deploying models in continuous integration, continuous delivery, and continuous training (CI/CD/CT) pipelines (e.g., Cloud Build)
Monitoring AI solutions12.9%
  • Building secure AI systems by protecting against unintentional exploitation and leaks of data or models (e.g., data exfiltration, malicious prompting, and sharing sensitive data with LLMs) using the appropriate security tool (e.g., Regex, safety filters, and Model Armor)
  • Aligning with responsible AI practices (e.g., monitoring for bias)
  • Model explainability on Agent Platform (e.g., Agent Platform Inference)
  • Configuring and using Model Monitoring on Gemini Enterprise Agent Platform to establish continuous evaluation metrics for production models
  • Monitoring for common issues (e.g., training-serving skew, data drift, concept drift, and feature attribution drift)
  • Monitoring, testing, and evaluating gen AI solutions

Domains

Architecting low-code AI solutions12.9%
Developing ML models using BigQuery ML or AutoML on Gemini Enterprise Agent Platform
  • Building models in BigQuery ML or Agent Platform AutoML (e.g., classification, regression, forecasting, and clustering) based on the business problem
  • Performing feature engineering or selection using BigQuery ML
  • Generating predictions using BigQuery ML
  • Training models using Agent Platform AutoML
  • Fine-tuning Gemini models using BigQuery
Building AI solutions using Google Cloud AI APIs or foundational models
  • Evaluating and selecting the appropriate model for a given task from Gemini Enterprise Agent Platform Model Garden
  • Building applications using industry-specific APIs (e.g., Document AI API, Vision API, and Translate API)
  • Building solutions and tuning models for specific use cases (e.g., Gemini, Imagen, Veo, and models as a service in Model Garden)
  • Optimizing Gemini-based applications for cost, latency, and availability
Collaborating within and across teams to manage data and models15.8%
Exploring and preprocessing data for ML
  • Organizing and exploring different data types (e.g., tabular, text, and images) for efficient experimenting, training, and serving
  • Choosing the right tool for data preprocessing based on scale and complexity (e.g., BigQuery [SQL], Dataflow, Apache Spark, and in-memory Python frameworks)
  • Creating and consolidating features in Gemini Enterprise Agent Platform Feature Store
  • Ensuring data privacy and handling sensitive information (e.g., personally identifiable information [PII])
Model prototyping using notebooks (e.g., Gemini Enterprise Agent Platform Workbench and Colab Enterprise)
  • Applying collaboration and security best practices when setting up and running notebook environments
  • Developing models in Agent Platform Workbench or Colab Enterprise notebooks using common frameworks (e.g., PyTorch, sklearn, and JAX)
  • Using a variety of foundational and open-source models in Model Garden to create model prototypes in notebook environments
Tracking and running ML experiments
  • Choosing the appropriate Google Cloud environment for development and experimentation (e.g., Experiments on Gemini Enterprise Agent Platform, Gemini Enterprise Agent Platform Pipelines, and Kubeflow Pipelines) given the framework
  • Evaluating predictive and gen AI solutions (e.g., model evaluation metrics and LLM-as-a-judge)
  • Tracking and comparing model artifacts, versions, and lineage (e.g., Experiments on Agent Platform and Gemini Enterprise Agent Platform ML Metadata)
Scaling prototypes into ML models20.8%
Building models given the task considering cost, complexity, latency, and scalability
  • Choosing the model type (e.g., ARIMA, DNN, and LLM)
  • Choosing the product (e.g., Agent Platform AutoML, BigQuery ML, and Agent Platform Pipelines)
  • Choosing the deployment strategy
  • Modeling techniques given interpretability requirements
Training models
  • Organizing training data (e.g., tabular, text, speech, images, and videos) on Google Cloud (e.g., Cloud Storage and BigQuery)
  • Ingesting structured and unstructured data from various sources into training pipelines
  • Model training using different software development kits (SDKs) (e.g., Agent Platform custom training, Kubeflow on Google Kubernetes Engine [GKE], Agent Platform AutoML, and Tabular Workflows) and organizing training on Google Cloud
  • Troubleshooting ML model training failures
  • Hyperparameter tuning
  • Fine-tuning foundational models from Agent Platform and Model Garden and when tuning should be considered
Choosing appropriate hardware for training
  • Evaluation of compute and accelerator options (e.g., CPU, GPU, and TPU)
  • Understanding the options for distributed training on GPUs and TPUs using data and model parallelism strategies
Serving and scaling models19.8%
Serving models
  • Deploying models for batch and online inference using appropriate services (e.g., Agent Platform, Model Garden, Cloud Run, and GKE)
  • Packaging and serving models from different frameworks (e.g., PyTorch and XGBoost) using prebuilt and custom containers
  • Organizing and versioning models in Gemini Enterprise Agent Platform Model Registry
  • Implementing model rollout strategies (e.g., A/B testing and canary deployments) to compare model versions
  • Developing solutions for inference preprocessing and postprocessing
Scaling online model serving
  • Managing and serving features using Agent Platform Feature Store
  • Deploying models to public and private endpoints
  • Choosing appropriate hardware (e.g., CPU, GPU, TPU, and edge)
  • Scaling the serving backend based on the throughput (e.g., Gemini Enterprise Agent Platform Inference and containerized serving)
  • Tuning ML models for training and serving in production
Automating and orchestrating ML pipelines17.8%
Developing end-to-end ML pipelines
  • Validating data and models
  • Building and orchestrating pipelines using managed or unmanaged services and from templates or custom solutions (e.g., Agent Platform Pipelines, Managed Service for Apache Airflow, and Ray on Gemini Enterprise Agent Platform)
  • Ensuring consistent data preprocessing between training and serving
Automating model retraining
  • Determining an appropriate retraining policy
  • Deploying models in continuous integration, continuous delivery, and continuous training (CI/CD/CT) pipelines (e.g., Cloud Build)
Monitoring AI solutions12.9%
Identifying risks to AI solutions
  • Building secure AI systems by protecting against unintentional exploitation and leaks of data or models (e.g., data exfiltration, malicious prompting, and sharing sensitive data with LLMs) using the appropriate security tool (e.g., Regex, safety filters, and Model Armor)
  • Aligning with responsible AI practices (e.g., monitoring for bias)
  • Model explainability on Agent Platform (e.g., Agent Platform Inference)
Monitoring, testing, and troubleshooting AI solutions
  • Configuring and using Model Monitoring on Gemini Enterprise Agent Platform to establish continuous evaluation metrics for production models
  • Monitoring for common issues (e.g., training-serving skew, data drift, concept drift, and feature attribution drift)
  • Monitoring, testing, and evaluating gen AI solutions