Google Professional Machine Learning Engineer Practice Test 1
Question 1
During a production ML inference release on Google Cloud, the following condition occurs: A document classification service sends huge media files through synchronous prediction calls whose processing exceeds clients' timeout limits. The service must meet its agreed quality and latency requirements. What implementation change resolves the specific issue without substituting an unrelated fix?
- Use asynchronous or batch processing with durable job status for oversized workloads, reserving online calls for bounded requests.
- Deploy a verified edge-compatible model for required offline decisions and sync summarized results when connectivity returns.
- Pin releases to reviewed immutable image digests and record image plus model artifact versions.
- Configure an appropriate request-concurrency or workload-driven scaling policy with resource and cooldown limits.
Explanation
Correct Answer
A:Use asynchronous or batch processing with durable job status for oversized workloads, reserving online calls for bounded requests. This is the targeted corrective response for online prediction timeout and batch fallback. Synchronous request limits cannot be solved merely by expanding the model context or retrying indefinitely. Verify the diagnosis using: Review payload size distributions, request duration, response timeout policies and tolerance for deferred completion. Confirm success using: Large documents finish without client timeout and consumers retrieve signed results using recorded job IDs.
Incorrect Answers
B:Deploy a verified edge-compatible model for required offline decisions and sync summarized results when connectivity returns. This action addresses edge versus cloud inference network constraints, not the failure in the described online prediction timeout and batch fallback case. The targeted correction is: Use asynchronous or batch processing with durable job status for oversized workloads, reserving online calls for bounded requests.
C:Pin releases to reviewed immutable image digests and record image plus model artifact versions. This action addresses model artifact image digest reproducibility, not the failure in the described online prediction timeout and batch fallback case. The targeted correction is: Use asynchronous or batch processing with durable job status for oversized workloads, reserving online calls for bounded requests.
D:Configure an appropriate request-concurrency or workload-driven scaling policy with resource and cooldown limits. This action addresses throughput-based autoscaling versus cpu metric, not the failure in the described online prediction timeout and batch fallback case. The targeted correction is: Use asynchronous or batch processing with durable job status for oversized workloads, reserving online calls for bounded requests.
Question 2
While preparing a production training workflow on Gemini Enterprise Agent Platform, the following condition occurs: An eight-GPU job scales only slightly beyond two GPUs because each small batch requires excessive cross-worker gradient synchronization. Engineers need an auditable, resource-aware correction to the issue. Which underlying engineering principle explains why the original approach fails?
- Reduced numeric precision narrows representable ranges and can underflow or overflow without proper scaling.
- Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated.
- Cloud accelerators have regional quota and stock limits distinct from an application's logical training configuration.
- Data parallel replicas each hold a model copy; they do not solve a model-too-large-for-one-device problem.
Explanation
Correct Answer
B:Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated. This explains the mechanism behind distributed gradient communication bottleneck and why the described failure is possible. A discriminating investigation is: Inspect all-reduce time, network bandwidth, effective batch size, worker stragglers and samples per second. The corresponding repair is: Tune supported distributed batch sizing, communication strategy or placement after measuring the bottleneck.
Incorrect Answers
A:Reduced numeric precision narrows representable ranges and can underflow or overflow without proper scaling. This principle describes mixed precision overflow diagnosis, which is not the cause of the observed distributed gradient communication bottleneck failure. The controlling explanation is: Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated.
C:Cloud accelerators have regional quota and stock limits distinct from an application's logical training configuration. This principle describes accelerator quota and regional availability, which is not the cause of the observed distributed gradient communication bottleneck failure. The controlling explanation is: Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated.
D:Data parallel replicas each hold a model copy; they do not solve a model-too-large-for-one-device problem. This principle describes model parallelism for memory-bound transformer, which is not the cause of the observed distributed gradient communication bottleneck failure. The controlling explanation is: Distributed speedups depend on communication-to-compute ratio, not merely the number of accelerators allocated.
Question 3
During a regulated MLOps pipeline deployment, the following condition occurs: A data ingestion pipeline changes a numeric feature into a string, but the model-serving path accepts it and silently substitutes missing values. The workflow must be repeatable and protect the last approved production model. Which result is the strongest postchange evidence of success?
- An incompatible payload fails visibly while compatible older versions follow documented conversion rules.
- One completed batch starts one intended pipeline run and unrelated object uploads do not trigger training.
- Staging and production runs report the same component digests and only documented configuration values differ.
- Changing the dataset version forces a rebuild while identical inputs may reuse validated cached results.
Explanation
Correct Answer
A:An incompatible payload fails visibly while compatible older versions follow documented conversion rules. This is specific acceptance evidence for serving schema validation component. Silent casting can hide producer errors and produce invalid model inputs across pipeline boundaries. It confirms the controlled correction: Enforce a shared typed schema contract and reject incompatible feature versions before training or inference.
Incorrect Answers
B:One completed batch starts one intended pipeline run and unrelated object uploads do not trigger training. This result could demonstrate success for eventbridge equivalent event trigger on storage but not for the stated serving schema validation component issue. The direct test is: An incompatible payload fails visibly while compatible older versions follow documented conversion rules.
C:Staging and production runs report the same component digests and only documented configuration values differ. This result could demonstrate success for pipeline environment parameter promotion but not for the stated serving schema validation component issue. The direct test is: An incompatible payload fails visibly while compatible older versions follow documented conversion rules.
D:Changing the dataset version forces a rebuild while identical inputs may reuse validated cached results. This result could demonstrate success for kubeflow component cache invalidation but not for the stated serving schema validation component issue. The direct test is: An incompatible payload fails visibly while compatible older versions follow documented conversion rules.
Question 4
During a controlled replay of the data and model handoff to a second engineering team, the following condition occurs: A project declares model B superior even though A was evaluated on last quarter and B on a deliberately easier random sample. The handoff must preserve privacy, lineage and testable behavior. Select TWO: the decisive diagnostic step and the targeted repair for this specific issue.
- Record input dataset versions, training runs, code revisions and output artifact links in the managed metadata and registry workflow.
- Inspect dataset and split fingerprints, label definitions, metric versions and evaluation timestamps for both experiment runs.
- Register a common versioned holdout and comparable metrics in Agent Platform Experiments before ranking candidates.
- Redact or tokenize unnecessary identifying fields before labeling and limit access to approved mappings.
- Inspect judge rubric prompts, evaluation dataset balance, repeated ratings, human benchmark and uncertainty in pairwise preferences.
Explanation
Correct Answers
B:Inspect dataset and split fingerprints, label definitions, metric versions and evaluation timestamps for both experiment runs. This inspects evidence relevant to experiment metric comparison with mismatched splits. Metrics from different data distributions are not directly comparable evidence of model improvement. After identifying the cause, the matching correction is: Register a common versioned holdout and comparable metrics in Agent Platform Experiments before ranking candidates.
C:Register a common versioned holdout and comparable metrics in Agent Platform Experiments before ranking candidates. This is the targeted corrective response for experiment metric comparison with mismatched splits. Metrics from different data distributions are not directly comparable evidence of model improvement. Verify the diagnosis using: Inspect dataset and split fingerprints, label definitions, metric versions and evaluation timestamps for both experiment runs. Confirm success using: Candidate scores can be reproduced on the same held-out data and their differences are attributable to model changes.
Incorrect Answers
A:Record input dataset versions, training runs, code revisions and output artifact links in the managed metadata and registry workflow. This is an action or diagnostic relevant to ml metadata lineage for production model, not one of the required two steps. In this scenario verify: Inspect dataset and split fingerprints, label definitions, metric versions and evaluation timestamps for both experiment runs. Then correct: Register a common versioned holdout and comparable metrics in Agent Platform Experiments before ranking candidates.
D:Redact or tokenize unnecessary identifying fields before labeling and limit access to approved mappings. This is an action or diagnostic relevant to data minimization for labeling inputs, not one of the required two steps. In this scenario verify: Inspect dataset and split fingerprints, label definitions, metric versions and evaluation timestamps for both experiment runs. Then correct: Register a common versioned holdout and comparable metrics in Agent Platform Experiments before ranking candidates.
E:Inspect judge rubric prompts, evaluation dataset balance, repeated ratings, human benchmark and uncertainty in pairwise preferences. This is an action or diagnostic relevant to llm-as-a-judge rubric reliability, not one of the required two steps. In this scenario verify: Inspect dataset and split fingerprints, label definitions, metric versions and evaluation timestamps for both experiment runs. Then correct: Register a common versioned holdout and comparable metrics in Agent Platform Experiments before ranking candidates.
Question 5
During a staged restoration test at a secondary serving deployment, the following condition occurs: A real-time scoring service reads a managed feature view that has not synchronized since yesterday while the source BigQuery table contains recent transaction updates. The team must prove an actual user-visible improvement without weakening access controls. What should the engineer inspect before selecting a correction?
- Examine effective IAM bindings, invocation audit logs and the intended partner service account identity.
- Inspect resolved endpoint address, authorized network routing, service access policy and connection logs.
- Inspect feature-view sync status, latest source timestamps, entity ID and online lookup freshness.
- Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata.
Explanation
Correct Answer
C:Inspect feature-view sync status, latest source timestamps, entity ID and online lookup freshness. This inspects evidence relevant to feature store freshness and online sync. Online feature serving requires an up-to-date synchronized view; source-table changes need not appear instantly. After identifying the cause, the matching correction is: Schedule and monitor approved Feature Store synchronization at the needed frequency, with stale-feature handling.
Incorrect Answers
A:Examine effective IAM bindings, invocation audit logs and the intended partner service account identity. This examines public endpoint iam oversharing rather than the specific mechanism reported here. It is not the decisive first check for feature store freshness and online sync. The relevant evidence is: Inspect feature-view sync status, latest source timestamps, entity ID and online lookup freshness.
B:Inspect resolved endpoint address, authorized network routing, service access policy and connection logs. This examines private vpc routing failure for serving rather than the specific mechanism reported here. It is not the decisive first check for feature store freshness and online sync. The relevant evidence is: Inspect feature-view sync status, latest source timestamps, entity ID and online lookup freshness.
D:Review immutable artifact IDs, Model Registry versions, evaluation reports, approvals and deployed revision metadata. This examines model registry version provenance rather than the specific mechanism reported here. It is not the decisive first check for feature store freshness and online sync. The relevant evidence is: Inspect feature-view sync status, latest source timestamps, entity ID and online lookup freshness.
Question 6
During an independent acceptance rehearsal of a customer-facing AI capability, the following condition occurs: A reporting team exports millions of BigQuery rows to a laptop for daily scoring although a trained BigQuery ML model already exists in the dataset. The owners require measurable quality and controlled operating cost. Which targeted engineering change most directly addresses the failure?
- Use ML.PREDICT on the appropriate BigQuery ML model within a controlled SQL batch job, writing results to governed tables.
- Use a BigQuery ML LINEAR_REG or suitable regression model with a numeric target and evaluate continuous-error metrics on held-out records.
- Move deterministic feature transformations into the supported BigQuery ML TRANSFORM workflow and use the same model at inference.
- Use Cloud Translation API for supported text translation, then separately validate downstream classification against localized content.
Explanation
Correct Answer
A:Use ML.PREDICT on the appropriate BigQuery ML model within a controlled SQL batch job, writing results to governed tables. This is the targeted corrective response for bigquery ml sql batch prediction integration. In-database prediction avoids unnecessary bulk data transfer when model and features are already in BigQuery. Verify the diagnosis using: Review row volume, model location, permissions and the current SQL report dependency on the local scoring script. Confirm success using: The scheduled report processes the full authorized partition and stores traceable predictions without a local export.
Incorrect Answers
B:Use a BigQuery ML LINEAR_REG or suitable regression model with a numeric target and evaluate continuous-error metrics on held-out records. This action addresses bigquery ml linear regression for continuous value, not the failure in the described bigquery ml sql batch prediction integration case. The targeted correction is: Use ML.PREDICT on the appropriate BigQuery ML model within a controlled SQL batch job, writing results to governed tables.
C:Move deterministic feature transformations into the supported BigQuery ML TRANSFORM workflow and use the same model at inference. This action addresses bigquery ml transform serving parity, not the failure in the described bigquery ml sql batch prediction integration case. The targeted correction is: Use ML.PREDICT on the appropriate BigQuery ML model within a controlled SQL batch job, writing results to governed tables.
D:Use Cloud Translation API for supported text translation, then separately validate downstream classification against localized content. This action addresses translate api for multilingual product descriptions, not the failure in the described bigquery ml sql batch prediction integration case. The targeted correction is: Use ML.PREDICT on the appropriate BigQuery ML model within a controlled SQL batch job, writing results to governed tables.
Question 7
During a production ML inference release on Google Cloud, the following condition occurs: A lightweight CPU model needs a custom HTTP handler with sporadic traffic, but the team provisions accelerator-bound infrastructure despite no accelerator operations. The service must meet its agreed quality and latency requirements. What acceptance result verifies the intended behavior after remediation?
- A known prediction request succeeds through the deployed endpoint and readiness reflects successful initialization.
- The selected service meets p95 latency and payload requirements with lower idle cost and reliable scaling.
- Identical sample text yields matching token IDs and model outputs in offline and online environments.
- Authorized lookups return expected recent values and freshness alarms fire when synchronization falls behind.
Explanation
Correct Answer
B:The selected service meets p95 latency and payload requirements with lower idle cost and reliable scaling. This is specific acceptance evidence for cloud run versus managed inference endpoint. Containerized CPU inference can be appropriate on serverless compute; specialized ML endpoints are not required for every model. It confirms the controlled correction: Benchmark a Cloud Run custom container against a managed Agent Platform endpoint and choose for workload fit.
Incorrect Answers
A:A known prediction request succeeds through the deployed endpoint and readiness reflects successful initialization. This result could demonstrate success for inference container health interface mismatch but not for the stated cloud run versus managed inference endpoint issue. The direct test is: The selected service meets p95 latency and payload requirements with lower idle cost and reliable scaling.
C:Identical sample text yields matching token IDs and model outputs in offline and online environments. This result could demonstrate success for inference preprocessing tokenization mismatch but not for the stated cloud run versus managed inference endpoint issue. The direct test is: The selected service meets p95 latency and payload requirements with lower idle cost and reliable scaling.
D:Authorized lookups return expected recent values and freshness alarms fire when synchronization falls behind. This result could demonstrate success for feature store freshness and online sync but not for the stated cloud run versus managed inference endpoint issue. The direct test is: The selected service meets p95 latency and payload requirements with lower idle cost and reliable scaling.
Question 8
During a staged restoration test at a secondary serving deployment, the following condition occurs: A text embedding model cannot fit into a chosen tiny GPU's memory, but a team only considers increasing container CPU cores. The team must prove an actual user-visible improvement without weakening access controls. Select TWO: the decisive diagnostic step and the targeted repair for this specific issue.
- Choose a GPU configuration with sufficient memory or validated model compression while benchmarking cost per query.
- Restrict invoke access to the necessary principal and endpoint, using appropriate short-lived authentication.
- Pin releases to reviewed immutable image digests and record image plus model artifact versions.
- Inspect model weights, KV or activation memory, concurrency load, GPU memory, and supported device types.
- Inspect batch size, kernel utilization, per-request overhead, queue wait and latency percentiles.
Explanation
Correct Answers
A:Choose a GPU configuration with sufficient memory or validated model compression while benchmarking cost per query. This is the targeted corrective response for gpu cost and memory fit for large embedding model. CPU cores do not expand device memory; accelerator selection must match model footprint and request shape. Verify the diagnosis using: Inspect model weights, KV or activation memory, concurrency load, GPU memory, and supported device types. Confirm success using: The model loads at target concurrency without OOM and meets latency and cost targets.
D:Inspect model weights, KV or activation memory, concurrency load, GPU memory, and supported device types. This inspects evidence relevant to gpu cost and memory fit for large embedding model. CPU cores do not expand device memory; accelerator selection must match model footprint and request shape. After identifying the cause, the matching correction is: Choose a GPU configuration with sufficient memory or validated model compression while benchmarking cost per query.
Incorrect Answers
B:Restrict invoke access to the necessary principal and endpoint, using appropriate short-lived authentication. This is an action or diagnostic relevant to public endpoint iam oversharing, not one of the required two steps. In this scenario verify: Inspect model weights, KV or activation memory, concurrency load, GPU memory, and supported device types. Then correct: Choose a GPU configuration with sufficient memory or validated model compression while benchmarking cost per query.
C:Pin releases to reviewed immutable image digests and record image plus model artifact versions. This is an action or diagnostic relevant to model artifact image digest reproducibility, not one of the required two steps. In this scenario verify: Inspect model weights, KV or activation memory, concurrency load, GPU memory, and supported device types. Then correct: Choose a GPU configuration with sufficient memory or validated model compression while benchmarking cost per query.
E:Inspect batch size, kernel utilization, per-request overhead, queue wait and latency percentiles. This is an action or diagnostic relevant to dynamic batch serving for gpu saturation, not one of the required two steps. In this scenario verify: Inspect model weights, KV or activation memory, concurrency load, GPU memory, and supported device types. Then correct: Choose a GPU configuration with sufficient memory or validated model compression while benchmarking cost per query.
Question 9
During a controlled replay of the data and model handoff to a second engineering team, the following condition occurs: A daily feature query scans two years of sales even though it uses only yesterday's records because the filter wraps the partition date in an incompatible expression. The handoff must preserve privacy, lineage and testable behavior. Which TWO complementary choices jointly establish the evidence and appropriate correction? Select TWO.
- Rewrite the filter to permit effective partition pruning while preserving its requested date boundaries.
- Review notebook execution identity, effective IAM roles, dataset access policies and audit records for protected tables.
- Use Agent Platform Experiments and pipeline metadata to track each trial's parameters, measurements and output artifacts.
- Standardize stable entity identifiers and validated feature definitions across training and online serving pathways.
- Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition.
Explanation
Correct Answers
A:Rewrite the filter to permit effective partition pruning while preserving its requested date boundaries. This is the targeted corrective response for bigquery partition filter cost optimization. Partition pruning avoids irrelevant I/O when predicates match the storage partitioning scheme. Verify the diagnosis using: Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition. Confirm success using: The equivalent daily query scans only the eligible partitions and returns the same number of matched entities.
E:Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition. This inspects evidence relevant to bigquery partition filter cost optimization. Partition pruning avoids irrelevant I/O when predicates match the storage partitioning scheme. After identifying the cause, the matching correction is: Rewrite the filter to permit effective partition pruning while preserving its requested date boundaries.
Incorrect Answers
B:Review notebook execution identity, effective IAM roles, dataset access policies and audit records for protected tables. This is an action or diagnostic relevant to colab enterprise notebook data access scope, not one of the required two steps. In this scenario verify: Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition. Then correct: Rewrite the filter to permit effective partition pruning while preserving its requested date boundaries.
C:Use Agent Platform Experiments and pipeline metadata to track each trial's parameters, measurements and output artifacts. This is an action or diagnostic relevant to pipeline experiment versus notebook-only tracking, not one of the required two steps. In this scenario verify: Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition. Then correct: Rewrite the filter to permit effective partition pruning while preserving its requested date boundaries.
D:Standardize stable entity identifiers and validated feature definitions across training and online serving pathways. This is an action or diagnostic relevant to feature store online and offline entity alignment, not one of the required two steps. In this scenario verify: Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition. Then correct: Rewrite the filter to permit effective partition pruning while preserving its requested date boundaries.
Question 10
Within a jointly maintained Google Cloud data science workspace, the following condition occurs: A daily feature query scans two years of sales even though it uses only yesterday's records because the filter wraps the partition date in an incompatible expression. Collaborators must avoid expanding data access or breaking experiment reproducibility. What should the engineer inspect before selecting a correction?
- Inspect feature event timestamps, prediction cutoffs, feature joins and any post-cancellation administrative flags.
- Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition.
- Inspect join-key frequency distributions, per-task runtime, shuffle spill and the output cardinality before optimization.
- Restart the runtime, run the notebook top to bottom and compare variable initialization with committed source revision.
Explanation
Correct Answer
B:Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition. This inspects evidence relevant to bigquery partition filter cost optimization. Partition pruning avoids irrelevant I/O when predicates match the storage partitioning scheme. After identifying the cause, the matching correction is: Rewrite the filter to permit effective partition pruning while preserving its requested date boundaries.
Incorrect Answers
A:Inspect feature event timestamps, prediction cutoffs, feature joins and any post-cancellation administrative flags. This examines point-in-time feature leakage prevention rather than the specific mechanism reported here. It is not the decisive first check for bigquery partition filter cost optimization. The relevant evidence is: Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition.
C:Inspect join-key frequency distributions, per-task runtime, shuffle spill and the output cardinality before optimization. This examines managed spark join-key skew repair rather than the specific mechanism reported here. It is not the decisive first check for bigquery partition filter cost optimization. The relevant evidence is: Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition.
D:Restart the runtime, run the notebook top to bottom and compare variable initialization with committed source revision. This examines notebook transient state and execution order rather than the specific mechanism reported here. It is not the decisive first check for bigquery partition filter cost optimization. The relevant evidence is: Examine the partitioning column, query plan, bytes scanned and predicate form against the table's partition definition.