Google Cloud Data Engineer Certification

Navigate directories and manage local files using basic command-line operations.

Write a Python script that reads and transforms records from a CSV file.

Parse nested JSON data into tabular records with Python.

Handle a failed file or record operation using Python exceptions and logging.

Track a data script with Git commits and a feature branch.

Explain how Google Cloud organizations, folders, projects, regions, and zones relate.

Create a Google Cloud project with billing alerts and a controlled budget.

Authenticate with Google Cloud and inspect resources using Cloud Shell and the gcloud CLI.

Translate a workload's data volume, latency, consistency, and access-pattern requirements into storage criteria.

Choose between object, relational, document, wide-column, globally distributed relational, and in-memory storage models.

Select among Cloud Storage, Cloud SQL, AlloyDB, Firestore, Bigtable, Spanner, and Memorystore for a stated workload.

Compare ACID transaction requirements with availability and scalability trade-offs.

Normalize a transactional relational model to reduce update anomalies.

Design a dimensional model with fact and dimension tables for an analytics workload.

Choose data retention, archival, and deletion rules from business requirements.

Estimate and compare the main storage and access cost drivers for candidate services.

Create BigQuery datasets and tables with appropriate locations and schemas.

Load batch data into BigQuery from local files and Cloud Storage.

Query nested and repeated BigQuery fields using UNNEST.

Partition and cluster a BigQuery table for a stated query pattern.

Diagnose a costly or slow BigQuery query using its execution details.

Reduce BigQuery scan cost by rewriting a query and using partition pruning.

Choose between logical views, materialized views, and BI Engine for an analytics requirement.

Choose between on-demand pricing, BigQuery Editions, and reservations for a workload pattern.

Configure Cloud Storage classes and lifecycle rules for a raw data zone.

Query data in Cloud Storage through BigQuery external tables or BigLake tables.

Organize raw, curated, and serving zones for a data lake.

Register and discover data assets with Dataplex Catalog metadata.

Profile a dataset to identify schema, quality, and sensitivity issues.

Define executable data-quality checks for completeness, uniqueness, validity, and freshness.

Apply a federated governance model to data owned by multiple teams.

Select BigQuery, Cloud Storage, BigLake, and Dataplex components for a governed data-platform design.

Define sources, sinks, transformations, schedules, and service-level requirements for a batch pipeline.

Choose ETL or ELT for a pipeline based on scale, governance, and transformation needs.

Model a batch transformation with Apache Beam PCollections and transforms.

Run a basic Apache Beam batch pipeline on Dataflow.

Implement a repeatable SQL transformation and assertion with Dataform.

Configure a visual batch pipeline with Cloud Data Fusion for a low-code use case.

Run a basic Spark transformation on Dataproc.

Select among BigQuery, Dataflow, Data Fusion, and Dataproc for a batch-processing scenario.

Distinguish bounded batch data from unbounded streaming data.

Configure Pub/Sub topics, subscriptions, acknowledgments, and retention for an event source.

Design an event schema that supports versioning and deduplication.

Apply fixed windows and event-time timestamps in an Apache Beam pipeline.

Explain how watermarks, triggers, and allowed lateness affect streaming results.

Handle duplicate and late-arriving events with keys, windows, and idempotent writes.

Run a basic Pub/Sub-to-BigQuery streaming pipeline on Dataflow.

Choose between Pub/Sub, Apache Kafka, and batch ingestion for a stated requirement.

Represent pipeline dependencies and retries as a directed acyclic graph.

Create a basic Cloud Composer DAG that runs a scheduled data task.

Use Workflows to coordinate calls to managed Google Cloud services.

Choose between Cloud Composer, Workflows, and service-native scheduling for an orchestration scenario.

Separate development and production data workloads using projects, identities, and configuration.

Create a CI check that validates a SQL or Python data transformation before deployment.

Design an idempotent pipeline step that can be retried safely.

Define a deployment and rollback sequence for a data-pipeline change.

Assign predefined IAM roles according to least privilege for a data-pipeline scenario.

Configure a service account for a workload without using a long-lived user credential.

Design project, dataset, and table boundaries for team isolation and governance.

Choose between Google-managed encryption, customer-managed keys, and application-level encryption.

Apply BigQuery row-level security, column-level access controls, or masking to a stated requirement.

Detect and classify personally identifiable information using Sensitive Data Protection.

Choose a data location that satisfies residency and sovereignty constraints.

Use audit logs and data lineage to investigate who accessed or changed a data asset.

Define a service-level indicator and service-level objective for a data pipeline.

Create logs, metrics, dashboards, and alerts for a failed or delayed pipeline.

Diagnose a pipeline failure caused by permissions, quotas, malformed data, or unavailable resources.

Use BigQuery administration and execution information to identify workload bottlenecks.

Select BigQuery capacity and workload-management controls for interactive and batch priorities.

Estimate the effect of storage, compute, network, and orchestration choices on monthly cost.

Derive recovery time and recovery point objectives from business impact requirements.

Design a restart, replay, replication, or failover strategy for a stated failure scenario.

Inventory source systems, stakeholders, dependencies, and risks for a cloud data migration.

Select BigQuery Data Transfer Service, Database Migration Service, Datastream, Transfer Appliance, or network transfer for a migration scenario.

Define reconciliation checks that validate record counts, aggregates, and data fidelity after migration.

Plan a phased cutover and rollback for a data-platform migration.

Prepare a BigQuery dataset for secure use by a visualization tool.

Publish a governed dataset through Analytics Hub or authorized BigQuery sharing.

Train and evaluate a basic predictive model with BigQuery ML.

Prepare unstructured content, metadata, and embeddings for a retrieval-augmented generation dataset.

Convert a business scenario into functional, security, reliability, latency, and cost requirements.

Create a Google Cloud architecture diagram that maps each requirement to a justified service choice.

Build a small ingestion component that lands source data in the portfolio platform.

Implement and test one curated transformation for the portfolio platform.

Apply least-privilege access and sensitive-data controls to the portfolio platform.

Add data-quality checks and operational alerts to the portfolio platform.

Write a concise technical README explaining architecture, deployment, costs, and trade-offs.

Answer timed case-based questions by eliminating options that violate stated requirements.