support@Free-Braindumps.com
Register Login
Cisco
CompTIA
ISC
EC-Council
EXIN
Fortinet
ITIL
Juniper
Microsoft
VMware
Avaya
SAP
Google
NVIDIA
Salesforce
Amazon
Palo Alto Networks
ISC2
CrowdStrike
Anthropic
All Vendors
  1. Home
  2. Exams
  3. Google
  4. PROFESSIONAL MACHINE LEARNING ENGINEER

Free PROFESSIONAL MACHINE LEARNING ENGINEER Exam Braindumps
Professional Machine Learning Engineer Questions 106–110 (page: 23)

Updated On: 25-Oct-2026
Page 23 of 82
PDF with 402 Questions

You work at a bank. You need to develop a credit risk model to support loan application decisions. You decide to implement the model by using a neural network in TensorFlow. Due to regulatory requirements, you need to be able to explain the model’s predictions based on its features.
When the model is deployed, you also want to monitor the model’s performance over time. You decided to use Vertex AI for both model development and deployment.
What should you do?

  1. Use Vertex Explainable AI with the sampled Shapley method, and enable Vertex AI Model Monitoring to check for feature distribution drift.
  2. Use Vertex Explainable AI with the sampled Shapley method, and enable Vertex AI Model Monitoring to check for feature distribution skew.
  3. Use Vertex Explainable AI with the XRAI method, and enable Vertex AI Model Monitoring to check for feature distribution drift.
  4. Use Vertex Explainable AI with the XRAI method, and enable Vertex AI Model Monitoring to check for feature distribution skew.

Answer(s): A



You are an MLOps engineer at a financial services company with a mandate to modernize your machine learning infrastructure. Your team supports multiple data science groups that use a wide range of ML frameworks including scikit-learn, PyTorch, and custom C++ libraries. Your team also has complex, multi-step pipelines and requires full control over their containerized environments, including specific OS-level dependencies.
You need to build and manage a flexible training platform that can orchestrate these diverse, container-based ML workflows while giving you complete control over the underlying compute infrastructure. The platform must also adhere to the company's open-source portability standard.
What should you do?

  1. Use the Agent Platform SDK to launch custom training jobs. Configure each job to pull the appropriate pre-built container image for scikit-learn or PyTorch.
  2. Deploy Kubeflow on a Google Kubernetes Engine (GKE) cluster. Use the Kubeflow Pipelines (KFP) SDK to define and orchestrate the container-based workflows.
  3. Use Agent Platform AutoML to train, evaluate, and deploy all models.
  4. Execute BigQuery ML CREATE MODEL statements within scheduled queries to initiate training jobs directly where the source data resides.

Answer(s): B

Explanation:

Deploying Kubeflow on Google Kubernetes Engine (GKE) provides a flexible, open-source machine learning platform with full control over the underlying compute infrastructure and containerized environments. Kubeflow Pipelines (KFP) enables the orchestration of complex, multi-step ML workflows using custom containers that support diverse frameworks, including scikit-learn, PyTorch, and custom C++ libraries. GKE allows teams to configure compute resources, operating system dependencies, networking, and resource allocation according to their requirements. Because Kubeflow is an open-source, Kubernetes-native platform, it supports portability across compatible Kubernetes environments while providing the customization and scalability needed for enterprise MLOps.



You work for a healthcare organization and need to automate the patient triage process. You want to build a telephony agentic solution using the Gemini Live API to perform human-like conversations with patients and gather necessary patient information. You plan to pilot the solution in Japanese. You need to minimize latency and development time, and ensure high, native-like fluency in the model’s responses.
What should you do?

  1. Use Gemini Live API Speech-to-Text functionality to capture the user’s input followed by Text-to-Speech streaming processing with Chirp to communicate with the patient in their language.
  2. Use the Gemini Live API, and configure the language_code field to ja-JP. Include a clear system instruction to respond in the patient’s language.
  3. Fine-tune the Gemini Flash model on a dataset of Japanese medical transcripts to ensure native fluency. Deploy this custom model behind the Gemini Live API.
  4. Use the Gemini Pro model to automatically select the appropriate language.

Answer(s): B

Explanation:

The Gemini Live API provides native, real-time audio interaction without a separate speech-to-text and text-to-speech pipeline, minimizing conversational latency and development effort. Configuring Japanese as the language and adding a clear system instruction guides the model to maintain fluent, natural Japanese throughout the patient interaction.


Reference:

https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/live-api https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/live-api/configure-language-voice https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/2-5-flash-live-api



You are optimizing the training of a 175-billion parameter LLM on Gemini Enterprise Agent Platform. You have provisioned a TPU v5p Pod slice with 32 chips for the job. You initially ran the training using standard data parallelism, but the job immediately failed with an out-of-memory (OOM) error. You need to implement a training strategy that resolves the memory issue and minimizes training latency.
What should you do?

  1. Configure the training to use a parameter server architecture where the model weights are stored in the host CPU memory.
  2. Enable gradient checkpointing, and continue using the data parallelism strategy.
  3. Configure the training to shard the model parameters and optimizer states, and continue using the data parallelism strategy.
  4. Scale up the TPU Pod slice to include more chips, and continue using the data parallelism strategy.

Answer(s): C

Explanation:

Sharding the model parameters and optimizer states across the TPU chips distributes the memory footprint instead of replicating the full 175-billion-parameter model on every device. This resolves the out-of-memory failure while retaining data parallelism for high-throughput training and lower overall training latency.



You are developing an ML model to predict house prices.
While preparing the data, you discover that an important predictor variable, distance from the closest school, is often missing and does not have high variance. Every instance (row) in your data is important. How should you handle the missing data?

  1. Delete the rows that have missing values.
  2. Apply feature crossing with another column that does not have missing values.
  3. Predict the missing values using linear regression.
  4. Replace the missing values with zeros.

Answer(s): C



  PDF
Viewing page 23 of 82
Viewing questions 111 - 115 out of 402 questions


Post your Comments and Discuss Google PROFESSIONAL MACHINE LEARNING ENGINEER exam prep with other Community members:

AI Tutor AI Tutor 👋 I’m here to help!