A.  

You will receive customer emails and need to extract date, sender email, and order ID. You should return the date, sender email, and order ID information in JSON format.

B.  

You will receive customer emails and need to extract date, sender email, and order ID. Return the extracted information in JSON format.

Here’s an example: {“date”: “April 16, 2024”, “sender_email”: “sarah.lee925@gmail.com”, “order_id”: “RE987D”}

C.  

You will receive customer emails and need to extract date, sender email, and order ID. Return the extracted information in a human-readable format.

D.  

You will receive customer emails and need to extract date, sender email, and order I

D.  

A.  

Use an access token belonging to service principals

B.  

Use a frequently rotated access token belonging to either a workspace user or a service principal

C.  

Use OAuth machine-to-machine authentication

D.  

Use an access token belonging to any workspace user

A.  

Host Llama Guard on Foundation Model API and use it to detect unsafe responses

B.  

Add some LLM calls to their chain to detect unsafe content before returning text

C.  

Add a regex expression on inputs and outputs to detect unsafe responses.

D.  

Ask users to report unsafe responses

A.  

Replace the current index with a Direct Vector Access index and push updates through a custom streaming job.

B.  

Update the pipeline mode to continuous mode rather than triggered mode.

C.  

Schedule a job to rebuild the index every hour after the ETL pipeline finishes.

D.  

Use a Delta Sync index so changes in the source Delta table are automatically synchronized to the index.

A.  

Slack integration to enable the agent in a dedicated workspace for the financial-analysis team.

B.  

Microsoft Teams integration to deploy the agent as a bot within Teams channels where analysts already collaborate.

C.  

REST API endpoint with a custom frontend application hosted on the company’s internal portal.

D.  

Databricks Apps to create a custom web application that analysts can access through their browsers with SSO authentication.

A.  

Implement a safety filter that detects any harmful inputs and ask the LLM to respond that it is unable to assist

B.  

Reduce the time that the users can interact with the LLM

C.  

Ask the LLM to remind the user that the input is malicious but continue the conversation with the user

D.  

Increase the amount of compute that powers the LLM to process input faster

A.  

MLflow Model Registry webhooks and a downstream audit log stored in an external database.

B.  

MLflow autologging and Delta Lake time travel on the inference table.

C.  

MLflow experiment tags and an automatically scripted changelog stored in a Databricks notebook.

D.  

MLflow Prompt Registry version history and logged runs that reference the prompt name and version used during inference.

A.  

Ingest documents from a source – > Index the documents and saves to Vector Search – > User submits queries against an LLM – > LLM retrieves relevant documents – > Evaluate model – > LLM generates a response – > Deploy it using Model Serving

B.  

Ingest documents from a source – > Index the documents and save to Vector Search – > User submits queries against an LLM – > LLM retrieves relevant documents – > LLM generates a response - > Evaluate model – > Deploy it using Model Serving

C.  

Ingest documents from a source – > Index the documents and save to Vector Search – > Evaluate model – > Deploy it using Model Serving

D.  

User submits queries against an LLM – > Ingest documents from a source – > Index the documents and save to Vector Search – > LLM retrieves relevant documents – > LLM generates a response – > Evaluate model – > Deploy it using Model Serving

A.  

flask

B.  

beautifulsoup

C.  

unstructured

D.  

numpy

A.  

Llama2-70b

B.  

BGE-large

C.  

MPT-7b

D.  

CodeLlama-34B

A.  

L!ama-2-70b-chat-hf

B.  

MPT-30B-lnstruct

C.  

DBRX

D.  

whisper-large-v3 (1.6B)

A.  

Limit the number of relevant documents available for the RAG application to retrieve from

B.  

Pick a smaller LLM that is domain-specific

C.  

Limit the number of queries a customer can send per day

D.  

Use the largest LLM possible because that gives the best performance for any general queries

A.  

context_sufficiency, correctness.

B.  

correctness, groundedness.

C.  

guideline_adherence, chunk_relevance.

D.  

relevance_to_query, chunk_relevance.

A.  

Compare the cosine similarities of the embeddings of returned results against those of a representative sample of test inputs

B.  

Compare the Bilingual Evaluation Understudy (BLEU) scores of returned results for a representative sample of test inputs

C.  

Compare the Recall-Onented-Understudy for Gistmg Evaluation (ROUGE) scores of returned results for a representative sample of test inputs

D.  

Compare the Levenshtein distances of returned results against a representative sample of test inputs

A.  

Ingest PDF documents about the monster truck team into a vector store and query it in a RAG architecture.

B.  

Write a system prompt for the agent listing available tools and bundle it into an agent system that runs a number of calls to solve a query.

C.  

Instruct the LLM to respond with “RAG”, “API”, or “TABLE” depending on the query, then use text parsing and conditional statements to resolve the query.

D.  

Build a system prompt with all possible event dates and table information in the system prompt. Use a RAG architecture to lookup generic text questions and otherwise leverage the information in the system prompt.

A.  

ROUGE metric

B.  

BLEU metric

C.  

NDCG metric

D.  

RECALL metric

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 LLM’s prediction function into a Flask application and serve using Gunicorn

A.  

Big Code Models Leaderboard

B.  

Number of model parameters

C.  

MTEB Leaderboard

D.  

Chatbot Arena Leaderboard

E.  

Number of model downloads last month

A.  

Use a managed Databricks MCP server integrated with Unity Catalog.

B.  

Use an external community MCP server for SQL and pass Unity Catalog tokens in prompts.

C.  

Run a custom MCP server on a self-managed VM that proxies the Unity Catalog API.

D.  

Expose JDBC directly to the model and enforce permissions in application code.

A.  

AI Gateway usage table.

B.  

System Serving, Served Entities table.

C.  

Inference Tables.

D.  

MLflow experiment run tables.

A.  

Add the section header as a prefix to chunks

B.  

Increase the document chunk size

C.  

Split the document by sentence

D.  

Use a larger embedding model

E.  

Fine tune the response generation model

A.  

It allows you to break down complex tasks into multiple independent subtasks. This enables the assistant to generate more comprehensive and accurate responses.

B.  

It allows you to reduce the latency of your applications. By having multiple chains participating in the response as a chain, you increase the rate at which the response is generated.

C.  

It allows you to decrease the effort involved in crafting a prompt. Chains make it possible to reuse prompt text across multiple different use cases.

D.  

It reduces the average cost of a typical request. Chains make more efficient use of the tokens produced to generate higher quality responses with fewer tokens.

A.  

Directly modify the LLM’s internal architecture to include preprocessing steps

B.  

It is better not to introduce custom code to preprocess prompts as the LLM has not been trained with examples of the preprocessed prompts

C.  

Rather than preprocessing prompts, it’s more effective to postprocess the LLM outputs to align the outputs to desired outcomes

D.  

Write a MLflow PyFunc model that has a separate function to process the prompts

A.  

DistilBERT

B.  

MPT-30B

C.  

Llama2-70B

D.  

DBRX

A.  

Multi-Agent Supervisor, because it can be used without Knowledge Assistant and Information Extraction agents.

B.  

Knowledge Assistant, because the chatbot interface can handle multi-turn conversations.

C.  

Knowledge Assistant, because it can be configured with multiple vector indexes to handle all three request types simultaneously.

D.  

Multi-Agent Supervisor, because it interprets incoming requests and delegates tasks to specialized sub-agents.

A.  

Tram the LLM to generate a single, comprehensive response without interacting with any external tools, relying solely on its pre-trained knowledge.

B.  

Implement a framework like ReAct which allows the LLM to generate reasoning traces and perform task-specific actions that leverage external tools if necessary.

C.  

Encourage the LLM to make multiple API calls in sequence without planning or structuring the calls, allowing the LLM to decide when and how to use external tools spontaneously.

D.  

Use a Chain-of-Thought (CoT) prompting technique to guide the LLM through a series of reasoning steps, then manually input the results from external tools for the final answer.

A.  

Implement AI guardrails to allow users to configure and enforce compliance

B.  

Leverage Databricks access control lists (ACLs) to configure permissions for accessing models

C.  

Use secure model features with Databricks Feature Store

D.  

Apply attribute-based access controls (ABAC) to limit unauthorized access