Study guide
1.2 From Vertex AI to Gemini Enterprise Agent Platform: The Product Name Map
Google Cloud Professional Machine Learning Engineer PMLE
Track your progress as you read
Finished sections get a check mark.
1 Chapter 1: Exam Overview & the Agent Platform Map
1.1 PMLE Exam Facts, Policies & the June 2026 Blueprint 1.2 From Vertex AI to Gemini Enterprise Agent Platform: The Product Name Map
2 Chapter 2: Low-Code ML with BigQuery ML
2.1 Choosing, Training & Predicting with BigQuery ML Models 2.2 Feature Engineering & Feature Selection in BigQuery ML 2.3 Fine-Tuning & Evaluating Gemini Models from BigQuery
3 Chapter 3: AutoML & Google Cloud AI APIs
3.1 Training Models with Agent Platform AutoML 3.2 Building with Document AI, Vision, Translation & Other AI APIs
4 Chapter 4: Foundation Models in Model Garden
4.1 Evaluating & Selecting Models in Model Garden 4.2 Building & Tuning Solutions with Gemini, Imagen, Veo & MaaS 4.3 Optimizing Gemini Applications for Cost, Latency & Availability
5 Chapter 5: Exploring, Preprocessing & Protecting Data
5.1 Organizing & Exploring Tabular, Text & Image Data 5.2 Choosing a Preprocessing Tool: BigQuery, Dataflow, Spark or Python 5.3 Creating & Consolidating Features in Agent Platform Feature Store 5.4 Protecting PII & Sensitive Data in ML Workflows
6 Chapter 6: Notebooks, Prototyping & Experiment Tracking
6.1 Workbench vs. Colab Enterprise: Secure, Collaborative Notebooks 6.2 Prototyping with PyTorch, scikit-learn, JAX & Model Garden in Notebooks 6.3 Experiment Environments, Tracking & Lineage with Experiments and ML Metadata
7 Chapter 7: Evaluating Predictive & Generative AI Solutions
7.1 Evaluation Metrics for Predictive Models 7.2 Evaluating Gen AI: Computation Metrics, Rubrics & LLM-as-a-Judge
8 Chapter 8: Choosing the Modeling Approach
8.1 Choosing the Model Type: ARIMA, Trees, DNNs & LLMs 8.2 Choosing the Product & Deployment Strategy 8.3 Modeling Techniques for Interpretability Requirements
9 Chapter 9: Training Data & Training SDKs
9.1 Organizing & Ingesting Training Data from Cloud Storage & BigQuery 9.2 Custom Training on Agent Platform: Containers, Jobs & Worker Pools 9.3 Kubeflow on GKE, AutoML & Tabular Workflows as Training Choices
10 Chapter 10: Troubleshooting, Hyperparameter Tuning & Fine-Tuning
10.1 Troubleshooting ML Training Failures 10.2 Hyperparameter Tuning on Agent Platform 10.3 Fine-Tuning Foundation Models & When to Tune
11 Chapter 11: Training Hardware & Distributed Training
11.1 Choosing CPUs, GPUs & TPUs for Training 11.2 Distributed Training: Data & Model Parallelism
12 Chapter 12: Deploying Models for Inference
12.1 Batch vs. Online Inference on Agent Platform 12.2 Serving Beyond Agent Platform Endpoints: Model Garden, Cloud Run & GKE 12.3 Prebuilt & Custom Serving Containers
13 Chapter 13: Model Versions, Rollouts & Inference Logic
13.1 Organizing & Versioning Models in Model Registry 13.2 Rollout Strategies: A/B Tests, Canaries & Traffic Splits 13.3 Inference Preprocessing & Postprocessing
14 Chapter 14: Scaling Online Serving
14.1 Serving Features Online with Feature Store 14.2 Public, Dedicated & Private Endpoints 14.3 Serving Hardware & Autoscaling for Throughput 14.4 Tuning Models for Production Training & Serving
15 Chapter 15: End-to-End ML Pipelines
15.1 Validating Data & Models in Pipelines 15.2 Agent Platform Pipelines with KFP: Components, Templates & Caching 15.3 Ensuring Consistent Preprocessing Between Training & Serving
16 Chapter 16: Orchestration with Managed Airflow & Ray
16.1 Orchestrating ML Workflows with Managed Service for Apache Airflow 16.2 Ray on Agent Platform for Distributed ML Workloads
17 Chapter 17: Automating Retraining & CI/CD/CT
17.1 Choosing a Retraining Policy 17.2 CI/CD/CT for ML with Cloud Build
18 Chapter 18: Securing AI Systems, Responsible AI & Explainability
18.1 Protecting AI Systems from Data & Model Exfiltration 18.2 Defending Gen AI Apps: Safety Filters, Model Armor & Regex Guardrails 18.3 Responsible AI: Detecting & Monitoring Bias 18.4 Model Explainability on Agent Platform
19 Chapter 19: Monitoring AI Solutions
19.1 Configuring Model Monitoring on Agent Platform 19.2 Training-Serving Skew, Data Drift, Concept Drift & Attribution Drift 19.3 Monitoring, Testing & Evaluating Gen AI Solutions in Production
1.2 From Vertex AI to Gemini Enterprise Agent Platform: The Product Name Map
Key Takeaways Explain This Topic AI
Google renamed Vertex AI to Gemini Enterprise Agent Platform (Agent Platform) in April 2026, and the PMLE exam guide uses the new names.
The API endpoint aiplatform.googleapis.com and the google-cloud-aiplatform Python SDK kept their names through the rebrand.
Cloud Composer is now Managed Service for Apache Airflow, and Dataproc is now Managed Service for Apache Spark.
Vertex Explainable AI was deprecated on March 16, 2026 and is scheduled to shut down on March 16, 2027.
AutoML Text and AutoML Video are retired; Google directs those use cases to Gemini prompting and tuning.
Last updated: September 2026
If you learned Google Cloud ML before 2026, many product names on the exam will look unfamiliar. On April 22, 2026 , Google's release notes announced that Vertex AI features now live in Gemini Enterprise Agent Platform , usually shortened to Agent Platform . Google described it as the evolution of Vertex AI: all Vertex AI services and future roadmap work are delivered through Agent Platform rather than a standalone service. The PMLE exam guide dated June 1, 2026 uses these new names throughout. This guide does too, with the former name in parentheses where that helps.
Core ML Name Changes
- Former Vertex AI name Agent Platform name
- Vertex AI Platform / Vertex AI Agent Platform
- Vertex AI Studio Agent Studio
- Vertex AI Model Garden Model Garden
- Vertex AI Models as a Service (MaaS) MaaS
- Vertex AI Training Agent Platform Managed Training
- Vertex AI Serverless Training Agent Platform Serverless Training
- Vertex AI Training Clusters Managed Training Clusters
- Vertex AI Prediction / Vertex AI Inference Agent Platform Inference
- Vertex AI Batch Inference / Online Inference Agent Platform Batch Inference / Online Inference
- Vertex AI Endpoints Agent Platform Endpoints
- Vertex AI Pipelines Agent Platform Pipelines
- Vertex AI Workbench / Colab Enterprise Agent Platform Workbench / Agent Platform Colab Enterprise
- Vertex AI Feature Store Agent Platform Feature Store
- Vertex AI Model Registry Agent Platform Model Registry
- Vertex AI Model Evaluation Agent Platform Model Evaluation
- Gen AI evaluation service Gen AI evals
- Vertex AI AutoML Agent Platform AutoML
- Vertex AI Experiments Experiments on Agent Platform
- Vertex AI Model Monitoring Model Monitoring on Agent Platform
- Ray on Vertex AI Ray on Agent Platform
- Vertex AI Forecasting Forecasting on Agent Platform
- Vertex AI Vector Search / Vector Search 2.0 Vector Search / Agent Retrieval
- Vertex AI Agent Engine Agent Runtime
- Vertex AI Search Agent Search
- What did not change
The REST endpoint is still aiplatform.googleapis.com . For example, a Gemini endpoint URL still looks like https://us-central1-aiplatform.googleapis.com/v1/projects/.../publishers/google/models/... .
The Python SDK package is still google-cloud-aiplatform , so code snippets on the exam may still import aiplatform .
Bookmarked console links redirect to the new Agent Platform navigation.
Features kept working. The rename is about branding and navigation, not a migration of your models or pipelines.
Two products named "Gemini Enterprise"
Gemini Enterprise is the workplace app that employees use to find information and run agents. Gemini Enterprise Agent Platform is the developer platform for models, training, tuning, serving, and agents. The PMLE exam is about the second one.
Data and Analytics Stack Renames
The exam page also mentions updates to Google Cloud's data and analytics stack. The renames most likely to appear in ML scenarios are:
Former name Current name Why it matters for ML
Cloud Composer Managed Service for Apache Airflow (Managed Airflow) Orchestrating ML workflows as Airflow DAGs (announced April 15, 2026)
Dataproc on Compute Engine and Serverless for Apache Spark Managed Service for Apache Spark Large-scale Spark preprocessing and feature engineering
Dataplex Universal Catalog Knowledge Catalog (renamed April 10, 2026) Discovering and governing datasets, including Agent Platform managed datasets
Cloud Data Loss Prevention Sensitive Data Protection (the DLP API keeps its name) Finding and de-identifying PII before training
Deprecations That Change the Right Answer
Renames don't change correct answers, but deprecations do. Know these:
Vertex Explainable AI was deprecated on March 16, 2026 . No new features are being added, and the APIs are scheduled to shut down on March 16, 2027 . Google suggests open-source libraries such as SHAP and LIME as alternatives. The exam still lists model explainability, so learn both the built-in attribution methods and the open-source path (Chapter 18).
Feature Store (Legacy) and Feature Store Optimized online serving were deprecated on February 17, 2026 , with shutdown on February 17, 2027. The current Feature Store uses BigQuery as the offline store and serves online through Bigtable online serving . Google directs embedding workloads to Vector Search (Chapters 5 and 14).
AutoML Text and AutoML Video are retired. Google now directs text classification, entity extraction, sentiment, and video use cases to Gemini prompting and supervised tuning . Agent Platform AutoML now covers image data (classification and object detection) and tabular data (classification, regression, and forecasting).
Workbench managed notebooks and user-managed notebooks reached end of life on March 30, 2026. Workbench instances and Colab Enterprise are the supported notebook environments (Chapter 6).
How to Read Older Practice Material
Much third-party material still says "Vertex AI." Translate names as you read, but check any design that depends on a retired capability. For example, a legacy practice answer might recommend AutoML Text for sentiment classification. On the current platform, the best low-code answer is a Gemini model called through Agent Platform, with supervised tuning if prompting falls short.
Use this three-step check whenever an older explanation names a product:
Translate the name. "Vertex AI Pipelines" becomes Agent Platform Pipelines, and "Vertex AI Endpoints" becomes Agent Platform Endpoints. The idea behind the answer usually holds.
Check the capability's status. If the answer needs AutoML Text, AutoML Video, legacy Feature Store entity types, managed notebooks, or long-term Explainable AI, look for the current replacement.
Re-apply the constraint. The current exam favors Google Cloud-native managed services. If a scenario asks for the least operational overhead, a managed Agent Platform or BigQuery capability usually beats a self-managed cluster running the same open-source tool.
Names that did not change
Some products you will see in ML scenarios kept their names: BigQuery , BigQuery ML , Cloud Storage , Dataflow , Pub/Sub , Cloud Run , Google Kubernetes Engine (GKE) , Cloud Build , Artifact Registry , Cloud Logging , Cloud Monitoring , and Model Armor . Gemini model families (Pro, Flash, and Flash-Lite), Imagen, and Veo are model names, not platform names. Their version numbers change often, so exam questions focus on choosing a model tier rather than memorizing version strings.