Snowflake
A cloud-native data warehouse platform — now the dominant data storage and querying layer for modern data teams.
Snowflake is the most widely adopted cloud data warehouse, offering scalable storage and compute separation that makes it practical for organizations of any size. It has become the default analytics database at companies building modern data stacks, displacing older on-premise solutions. Snowflake skills are required at most data engineering and analytics engineering roles.
Typical time to job-readiness: ~3 weeks.
Learning Snowflake
Beginner
Understand Snowflake's storage/compute separation, run queries in the web UI, and load data from flat files. Learn what virtual warehouses are and why they matter for cost.
Intermediate
Work with semi-structured data (VARIANT, FLATTEN), use time travel for data recovery, and optimize query performance with clustering keys.
Advanced
Multi-cluster warehouse configuration, data sharing across accounts, Snowpark for Python-based data engineering, and cost governance. SnowPro Core certification adds credibility.
Key concepts
- Separation of storage and compute — scale them independently and pay only when compute runs
- Virtual warehouses — clusters of compute resources that can be suspended when idle
- Time Travel — query historical data up to 90 days back, undo accidental changes
- Zero-copy cloning — instantly clone databases or tables without duplicating storage
- VARIANT data type for semi-structured JSON/Avro/Parquet stored natively
- Data sharing — share live data across accounts without copying or moving it
Common interview topics
- How does Snowflake's storage/compute separation differ from traditional databases
- What is a virtual warehouse and how does auto-suspend save money
- How does Time Travel work and what are its limitations
- How would you optimize a slow Snowflake query
- Explain micro-partitioning and clustering keys in Snowflake