AI & Data
Google Cloud Data Engineer: Dataflow vs Dataproc for Data Processing
Dataflow and Dataproc have historically represented two different ways to process large datasets on Google Cloud: one centered on the…
ExamTopics reading room
All articles
Explore more technical insights and certification explainers. Page 38 of 195.
From the editorial desk
Browse the archive
Technical explainers and certification perspectives from our editorial library.
Project Management & Governance
Google Cloud Data Engineer: Data Governance with Dataplex
Data governance in Google Cloud is no longer just a question of where a dataset lives or who has an…
AI & Data
Google Cloud Data Engineer: Reliable ELT Pipelines in BigQuery
ELT—extract, load, transform—fits BigQuery well because raw data can be loaded into the warehouse first and transformed with scalable SQL…
AI & Data
Google Cloud Data Engineer: BigQuery Partitioning & Clustering
Partitioning and clustering are two of the most important physical design decisions for large BigQuery tables because they influence how…
AI & Data
Google Cloud ML Engineer: Vertex AI Model Training and Deployment
Model training and deployment are the point where machine-learning ideas become operational services. On Google Cloud, the current Professional Machine…
AI & Data
Google Cloud ML Engineer: Responsible AI & Governance
Responsible AI is the discipline of building and operating AI systems in a way that accounts for safety, fairness, privacy,…
AI & Data
Google Cloud ML Engineer: Vertex AI Model Monitoring & Drift
A machine-learning model can remain technically available while becoming less useful every day. User behavior changes, upstream data pipelines evolve,…
AI & Data
Google Cloud ML Engineer: MLOps Pipelines
MLOps pipelines turn machine-learning work from a sequence of manual notebook steps into a repeatable production system. In Google Cloud,…
Cloud Computing
Google Cloud ML Engineer: Feature Engineering for ML
Feature engineering is the work of turning raw observations into model inputs that make a machine-learning problem learnable, stable, and…