Snowflake GES-C01 : SnowPro® Specialty: Gen AI Certification Exam

An analytics team needs to use various Snowflake Cortex LLM functions and wants to understand the cost implications and governance mechanisms for controlling model access and usage. The "ACCOUNTADMIN' has restricted general LLM access using the 'CORTEX_MODELS_ALLOWLIST' parameter. The team is particularly interested in how model complexity and specific function calls impact billing and how access can be managed effectively for different user groups.

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

Explanation: Only visible for Free4Torrent members. You can sign-up / login (it's free).

An operations manager is concerned about the governance and monitoring of a newly deployed Cortex Analyst application, especially regarding cost tracking and how multi-turn conversations are handled. Which statement accurately describes a feature for monitoring or managing a Cortex Analyst application?

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

Explanation: Only visible for Free4Torrent members. You can sign-up / login (it's free).

An organization operating in the AWS US West 2 (Oregon) region needs to process sensitive customer support tickets using Snowflake Cortex LLM functions. Due to the diverse availability of specific LLMs, they are considering enabling CORTEX_ENABLED_CROSS_REGION. What is a key data safety and security consideration when enabling CORTEX_ENABLED_CROSS_REGION for Snowflake Cortex LLM functions, specifically regarding data storage and persistence?

• A. It enables inference for features not supported in the local region by allowing data to be processed in a different Snowflake region, and user inputs, prompts, and outputs are not stored or cached.
• B. It may lead to increased compute costs if the cross-region model is more expensive, but data movement guarantees remain unchanged.
• C. User inputs and service-generated prompts will be stored in a cache in the remote region to optimize subsequent requests.
• D.
  
• E. Data transmitted across regions for inference is encrypted by default, but the encryption keys are managed by the third-party cloud provider in the remote region.

Explanation: Only visible for Free4Torrent members. You can sign-up / login (it's free).

An ML engineer is designing a Cortex Agent to provide highly accurate and contextualized responses. They intend for the agent to use state-of-the-art LLMs for orchestration and to maintain a specific brand tone in its outputs. Considering the available models and configurations for Cortex Agents, which statement is true?

• A. To ensure the agent's responses adhere to a desired brand and tone, 'Response instructions' can be configured, which guide the agent's output style and persona.
• B. Cortex Agents are restricted to using only Snowflake Arctic models for orchestration, due to security and governance requirements, and must operate in the account's default region.
• C. Cortex Agents primarily interact with data through the Snowflake Model Registry API to retrieve and update fine-tuned model parameters during their iterative planning phase.
• D. The agent's 'Planning' component is specifically responsible for evaluating results after each tool use and deciding the subsequent steps in the query resolution process, acting as a feedback loop.
• E. When an agent utilizes an LLM for orchestration, the system ensures that cross-region inference is automatically enabled without any latency implications, making region selection irrelevant.

Explanation: Only visible for Free4Torrent members. You can sign-up / login (it's free).

An operations manager is tasked with monitoring the cost and ensuring compliance for a Cortex Analyst deployment that uses the REST API. They are particularly concerned with accurately tracking credit consumption and understanding the implications of enabling external models. Which of the following statements correctly describe aspects of Cortex Analyst cost and governance?

• A. Enabling Azure OpenAI models via the ' ENABLE_CORTEX_ANALYST_MODEL_AZURE_OPENAF parameter ensures that all customer data and prompts remain within Snowflake's governance boundary and fully respect RBAC policies for those models.
• B. Snowflake's view provides granular usage information for REST API requests to cortex Analyst, including tokens processed per model.
• C. The view can be queried to track detailed information about cortex Analyst requests, including generated SQL and any errors.
• D. Credit consumption for Cortex Analyst is primarily based on the number of tokens processed by the underlying LLMs, with more complex natural language questions leading to higher token usage and costs.
• E. Cortex Analyst credit usage is based on the number of messages processed, at a rate of 67 Credits per 1 ,000 messages, and only successful responses (HTTP 200) are counted.

Explanation: Only visible for Free4Torrent members. You can sign-up / login (it's free).