How raghul-g2002/snowflake-introduction-dev works
Project type and purpose
• From the top-level README (“snowflake-introduction-dev”) and the directory names (introduction-m1, snowflake features-m2, snowflake-genai-m3, retrieval-augmented-generation-snowflake, cortex-analyst-with-cortex-search), this is a collection of Snowflake training/demo modules.
• Each folder corresponds to a lesson or “module” demonstrating Snowflake’s capabilities: basic ingestion and staging, SQL features, Snowpark/Python, GenAI, RAG, and Cortex integrations.
File structure
• introduction-m1: pure SQL scripts for database/schema creation, stages, ingestion, views.
• snowflake features-m2: SQL scripts plus Python (Snowpark) and a CLI snippet, showcasing advanced features (UDFs, RBAC, time travel, cloning, resource monitors).
• snowflake-genai-m3: SQL for Cortex LLM function + Snowpipe, Python for ML (XGBoost) and Streamlit.
• retrieval-augmented-generation-snowflake: Python Streamlit apps, SQL for RAG pipeline.
• cortex-analyst-with-cortex-search: raw_data_stage folder (CSV data), YAML config, SQL analytics and a Streamlit demo script using Cortex Search.
README for additional insights
• The top-level README only names the repo; sub-folders contain their own README.md (e.g., cortex-analyst-with-cortex-search/README.md) describing that demo in detail.