Free Databricks Databricks-Certified-Generative-AI-Engineer-Associate Exam Questions

Become Databricks Certified with updated Databricks-Certified-Generative-AI-Engineer-Associate exam questions and correct answers

Page: 1 / 30

Total 150 Questions | Updated On: Jan 08, 2025

Add To Cart

Question 1

As a Gen AI engineer working on batch inference with MLflow, which of the following methods are correctly used for performing batch inference in single-node and multi-node environments? Choose two correct answers.

A : Implement mlflow.pyfunc.spark_udf(spark, model_uri=...) and result_df = prod_df.withColumn("result", model_udf("input")) for multi-node batch inference.

B : Apply mlflow.pyfunc.spark_udf(spark, model_uri=...) and .predict() for single-node batch inference.

C : Utilize mlflow.pyfunc.load_model() and result_df = prod_df.withColumn("result", model_udf("input")) for multi-node batch processing.

D : Use mlflow.pyfunc.load_model() and .predict() for single-node batch inference.

Answer: A,D

Question 2

What is a key functionality of the background service built on Unity Catalog for managing data?

A : It allows for manual creation of dashboards and requires custom setup for each metric.

B : It provides a static report of metrics without real-time updates or customization options.

C : It performs incremental processing of data, calculates profile and drift metrics, auto-generates dashboards, and includes automatic PII detection.

D : It focuses exclusively on managing user access to data without processing or visualizing metrics.

Answer: C

Question 3

A Generative Al Engineer has already trained an LLM on Databricks and it is now ready to be deployed. Which of the following steps correctly outlines the easiest process for deploying a model on Databricks?

A : Log the model as a pickle object, upload the object to Unity Catalog Volume, register it to Unity Catalog using MLflow, and start a serving endpoint

B : Log the model using MLflow during training, directly register the model to Unity Catalog using the MLflow API, and start a serving endpoint

C : Save the model along with its dependencies in a local directory, build the Docker image, and run the Docker container

D : Wrap the LLMs prediction function into a Flask application and serve using Gunicorn

Answer: B

Question 4

Why summarizing retrieved documents before feeding them into an AI system is recommended?

A : To help the AI system to generate responses in multiple languages.

B : To eliminate irrelevant information, making the document shorter.

C : To allow the AI system to process multiple documents simultaneously.

D : To ensure the AI system stays within its context window and focuses on the most essential information.

Answer: D

Question 5

How can a Generative AI Engineer handle complex documents (with mixed text and images) for data extraction and chunking for a RAG based LLM application?

A : Using python libraries like PyPDF

B : Using Multi-modal models like OpenAI's GPT-4o and beyond

C : All of these options

D : Using layout model from Hugging Face

Answer: C

Page: 1 / 30

Total 150 Questions | Updated On: Jan 08, 2025

Add To Cart