Snowflake SnowPro Specialty: Gen AI (GES-C02) · 38% practice allocation
5 free sample questions from the Snowflake Cortex AI and LLM Functions domain, with answers and explanations. The full GES-C02 bank has 241 questions across 4 domains. Questions are scenario-based and written from the official Snowflake exam guide.
A pipeline must return the numeric sentiment of customer reviews as a score. Which Cortex function is purpose-built for this?
SENTIMENT returns a sentiment score for input text, purpose-built for this task. PARSE_DOCUMENT extracts text from files, EMBED_TEXT_1024 produces embeddings, and a bare COMPLETE would need careful prompting to match SENTIMENT's consistency.
Which older-style function name corresponds to AI_CLASSIFY?
CLASSIFY_TEXT is the older-style name paired with the current AI_CLASSIFY. SENTIMENT pairs with AI_SENTIMENT, and TRANSLATE and EXTRACT_ANSWER are unrelated task functions.
A request requires checking a customer's stated income against a policy document's eligibility thresholds AND computing their debt-to-income ratio from transaction tables, then producing a single approve/deny recommendation. What kind of Cortex capability does this multi-step, multi-tool reasoning call for?
This needs retrieval from a policy document, computation from transaction data, and synthesis into one recommendation — exactly a Cortex Agent's orchestration role. A single SUMMARIZE or AI_CLASSIFY call only handles one piece of the task, and fine-tuning doesn't provide the real-time cross-source reasoning needed here.
After using AI_FILTER to isolate support tickets mentioning a competitor by name, the team wants one summary paragraph describing the themes across all of those filtered tickets. Which function produces that single synthesized result across the group of rows?
AI_AGG is the aggregate-style function that applies an instruction across a whole group of rows and returns one synthesized result. AI_COMPLETE and SUMMARIZE called per row would each return separate per-row outputs, not one merged answer, and AI_CLASSIFY assigns labels rather than synthesizing narrative themes.
A data engineer needs to generate a one-off paraphrase of a marketing sentence using a chosen LLM inside a SQL SELECT. Which function is the general-purpose entry point for this?
AI_COMPLETE takes a model name and a prompt and returns generated text — the general-purpose generation entry point. AI_CLASSIFY assigns categories, EMBED_TEXT_768 produces vectors, and AI_FILTER returns a boolean, none of which generate free text.
The free GES-C02 practice test mixes about 20 questions weighted like the exam. The full 241-question bank adds instant grading, flashcards and a timed mock exam.