Introduction

1. Google Cloud Professional Machine Learning Engineer Certification
2. Google Cloud Professional ML Engineer Objective Map

Section 1: Framing ML Problems

1. Translating Business Use Cases
2. Machine Learning Approaches
3. ML Success Metrics
4. Responsible AI Practices
5. Summary
6. Exam Essentials
7. Review Questions

Section 2: Exploring Data and Building Data Pipelines

1. Visualization
2. Statistics Fundamentals
3. Data Quality and Reliability
4. Establishing Data Constraints
5. Running TFDV on Google Cloud Platform
6. Organizing and Optimizing Training Datasets
7. Handling Missing Data
8. Data Leakage
9. Summary
10. Exam Essentials
11. Review Questions

Section 3: Feature Engineering

1. Consistent Data Preprocessing
2. Encoding Structured Data Types
3. Class Imbalance
4. Feature Crosses
5. TensorFlow Transform
6. GCP Data and ETL Tools
7. Summary
8. Exam Essentials
9. Review Questions

Section 4: Choosing the Right ML Infrastructure

1. Pretrained vs. AutoML vs. Custom Models
2. Pretrained Models
3. AutoML
4. Custom Training
5. Provisioning for Predictions
6. Summary
7. Exam Essentials
8. Review Questions

Section 5: Architecting ML Solutions

1. Designing Reliable, Scalable, and Highly Available ML Solutions
2. Choosing an Appropriate ML Service
3. Data Collection and Data Management
4. Automation and Orchestration
5. Serving
6. Summary
7. Exam Essentials
8. Review Questions

Section 6: Building Secure ML Pipelines

1. Building Secure ML Systems
2. Identity and Access Management
3. Privacy Implications of Data Usage and Collection
4. Summary
5. Exam Essentials
6. Review Questions

Section 7: Model Building

1. Choice of Framework and Model Parallelism
2. Modeling Techniques
3. Transfer Learning
4. Semi‐supervised Learning
5. Data Augmentation
6. Model Generalization and Strategies to Handle Overfitting and Underfitting
7. Summary
8. Exam Essentials
9. Review Questions

Section 8: Model Training and Hyperparameter Tuning

1. Ingestion of Various File Types into Training
2. Developing Models in Vertex AI Workbench by Using Common Frameworks
3. Training a Model as a Job in Different Environments
4. Hyperparameter Tuning
5. Tracking Metrics During Training
6. Retraining/Redeployment Evaluation
7. Unit Testing for Model Training and Serving
8. Summary
9. Exam Essentials
10. Review Questions

Section 9: Model Explainability on Vertex AI

1. Model Explainability on Vertex AI
2. Summary
3. Exam Essentials
4. Review Questions

Section 10: Scaling Models in Production

1. Scaling Prediction Service
2. Serving (Online, Batch, and Caching)
3. Google Cloud Serving Options
4. Hosting Third‐Party Pipelines (MLflow) on Google Cloud
5. Testing for Target Performance
6. Configuring Triggers and Pipeline Schedules
7. Summary
8. Exam Essentials
9. Review Questions

Section 11: Designing ML Training Pipelines

1. Orchestration Frameworks
2. Identification of Components, Parameters, Triggers, and Compute Needs
3. System Design with Kubeflow/TFX
4. Hybrid or Multicloud Strategies
5. Summary
6. Exam Essentials
7. Review Questions

Section 12: Model Monitoring, Tracking, and Auditing Metadata

1. Model Monitoring
2. Model Monitoring on Vertex AI
3. Logging Strategy
4. Model and Dataset Lineage
5. Vertex AI Experiments
6. Vertex AI Debugging
7. Summary
8. Exam Essentials
9. Review Questions

Section 13: Maintaining ML Solutions

1. MLOps Maturity
2. Retraining and Versioning Models
3. Feature Store
4. Vertex AI Permissions Model
5. Common Training and Serving Errors
6. Summary
7. Exam Essentials
8. Review Questions

Section 14: BigQuery ML

1. BigQuery – Data Access
2. BigQuery ML Algorithms
3. Explainability in BigQuery ML
4. BigQuery ML vs. Vertex AI Tables
5. Interoperability with Vertex AI
6. BigQuery Design Patterns
7. Summary
8. Exam Essentials
9. Review Questions