An organization is implementing a two-tier LLM access control strategy in Snowflake. First, common models like 'mistral-7b' and 'llama3.1-8b' need to be broadly accessible to all users granted the 'SNOWFLAKE-CORTEX USER database role. Second, a specialized data science team, using the ANALYST ROLE', requires exclusive access to the higher-capability 'claude-3-5- sonnet' model, which should NOT be generally available through the broad access mechanism. Which set of SQL commands, executed by the 'ACCOUNTADMIN" role, correctly establishes this access control strategy?

• A.
  
• B.
  
• C.
  
• D.
  
• E.
  

A marketing team is analyzing social media comments using Snowflake and wants to categorize them into predefined campaign sentiments (e.g., 'Positive Campaign Engagement', 'Negative Campaign Feedback', 'Neutral Discussion'). They decide to use the SNOWFLAKE. CORTEX. CLASSIFY TEXT function for this task. Which of the following statements about its usage are correct?

• A. To provide more context and potentially improve classification accuracy, categories within the can be defined as SQL objects, including 'description' and 'examples' fields.
• B. The input string to CLASSIFY_TEXT is case-insensitive, meaning 'Great product!' and 'great product!' will yield identical classification results due to automatic normalization.
• C. If the input text exceeds a model-specific token limit, CLASSIFY_TEXT will automatically truncate the text before processing without raising an error.
• D. CLASSIFY_TEXT can return a JSON object with a 'label' field, where the value of this field indicates the classified category of the input text.
• E. The argument must contain exactly two string values for effective binary classification, otherwise an error is returned.

A data engineer is designing an automated pipeline in Snowflake to process streaming customer support tickets using LLM functions. To ensure the pipeline is robust against unexpected LLM errors (e.g., due to malformed input or transient service issues), they decide to use SNOWFLAKE.CORTEX.TRY_COMPLETE instead of SNOWF LAKE .CORTEX.COMPLETE
. Which of the following is the most significant advantage of using
TRY_COMPLETE
in this scenario?

• A. It processes tokens at a significantly reduced rate compared to
• B. It offers enhanced security features, such as automatic input sanitization, which are not available in the standard
• C. It provides a built-in retry mechanism for LLM calls that encounter temporary failures, automatically re-submitting the prompt up to three times.
• D. It returns
• E. It automatically adjusts the LLM's

A developer is instrumenting a RAG application using the TruLens SDK within Snowflake AI Observability. The application has distinct functions for retrieving context and generating a completion. To ensure clear tracing and readability, which span_type should ideally be used for the function responsible for retrieving relevant text from the vector store?

• A.
  
• B.
  
• C. No specific span_type is needed; the default instrumentation is sufficient for all functions.
• D.
  
• E.
  

A new data analyst is trying to incorporate sentiment analysis using SNOWFLAKE. CORTEX. SENTIMENT within a Snowflake data pipeline that uses dynamic tables. They execute the following SQL to create a dynamic table for daily sentiment aggregation:

However, this operation fails. Which of the following is the most direct reason for the failure of this specific setup?

• A. SNOWFLAKE. CORTEX. SENTIMENT and other Snowflake Cortex functions are currently incompatible with dynamic tables.
• B. The CORTEX_USER database role was not granted to the analyst's role, preventing the execution of Cortex functions.
• C. The review_content column, if containing non-English text, would cause the SENTIMENT function to fail outright rather than produce inaccurate results.
• D. The TARGET_LAG for dynamic tables must be explicitly set to '1 day' or longer when integrating with Cortex functions.
• E. The warehouse my_analytics_wh is likely not a Snowpark-optimized warehouse, which is a requirement for Cortex functions within dynamic tables.