Chapter 4 of 4 · Snowflake GES-C02
Snowflake Document Processing
GES-C02 document processing chapter overview: AI_PARSE_DOCUMENT modes, AI_EXTRACT, stages, streams and tasks, errors, cost and arctic-extract fine-tuning.
- 7topics
- 34 minfast track
- 3diagrams and exhibits
- 15%of the exam
Topics in this chapter
Take them in order, or jump to the one you need.
- 1AI_PARSE_DOCUMENT - Document Parsing Functions, OCR and LAYOUT ModesPick the right Snowflake parsing mode: AI_PARSE_DOCUMENT with OCR, LAYOUT or AGENTIC, TO_FILE inputs, output shape and the…5 min
- 2AI_PARSE_DOCUMENT - page_split, Page Selection and Page LimitsHandle pages in AI_PARSE_DOCUMENT: page_split output, page_filter and FL_FILTER_PAGES ranges, and the 2,000 page and 100 MB…Diagram5 min
- 3AI_EXTRACT - Response Format and Prompt EngineeringUse AI_EXTRACT to pull fields, lists and tables from documents: responseFormat forms, JSON schema, returned object, scores…Diagram5 min
- 4Preparing and Uploading Documents - Stages, Formats and Size LimitsPrepare documents for Snowflake Cortex: stages, directory tables, upload methods, supported formats, the 100 MB size cap, and…4 min
- 5Automated Document Pipelines - Orchestration with Streams and TasksAutomate document processing in Snowflake: directory tables, streams on stages, tasks with SYSTEM$STREAM_HAS_DATA, serverless…5 min
- 6Troubleshooting Document Processing - Errors, GET_PRESIGNED_URL and PrivilegesTroubleshoot document processing: AI_EXTRACT and AI_PARSE_DOCUMENT errors, cross-region behavior, GET_PRESIGNED_URL file checks…Diagram5 min
- 7Cost, Best Practices and Fine-Tuning arctic-extract ModelsControl document processing cost and accuracy: page and token billing, warehouse size, page filters, scores, and fine-tuning…5 min
About this chapter
What this chapter covers
Snowflake Document Processing is the chapter on turning files such as PDFs, scans and slide decks into text and structured data with Snowflake Cortex AI functions. For the SnowPro Specialty Gen AI GES-C02 exam it tests decisions: which function, which mode, which option, which privilege, and what to check when it fails.
Questions are scenarios that start with "A Gen AI Specialist needs to ...". The right answer usually turns on one detail, such as the default mode, an exclusive page index, a limit, or a setting that decides whether one bad file fails a whole batch.
How the topics fit together
Files land on a stage, a function reads them, and a pipeline repeats the work for each new file.
- Parsing covers the modes of AI_PARSE_DOCUMENT and how pages are split and selected.
- Extraction covers AI_EXTRACT, its response formats and how to write good questions.
- Preparation and pipelines cover stages, uploads, requirements, streams and tasks.
- Operations cover errors, GET_PRESIGNED_URL, privileges, cost and fine-tuning.
Parse when you need the whole content, extract when you need specific fields. A scenario that asks for "the invoice number and total" points to AI_EXTRACT.
How to use these pages
Each topic page has the same shape, so you can skim for what you need.
- What it is: the idea in plain words.
- How it works: the mechanism, with flows, SQL and comparison tables.
- Common mistakes: the classic traps.
- What to remember for the exam: the facts to memorize.
- Check yourself: three questions to test your understanding.
Read the pages in order the first time. Later pages reuse the stages and functions from the earlier ones.
Ready for more practice?
The full Snowflake GES-C02 question bank has 585 questions, with timed practice exams and an explanation for every answer.
This study guide is independent and is not affiliated with, endorsed by, or sponsored by Snowflake. Vendor names, certification names and exam codes are trademarks of their respective owners and are used for identification only.