Guides
Guides you can run, not just read.
Each guide answers one data quality or metadata question with complete code we ran before publishing, and the real output it produced. Download the file, run it on the platform and data named in its Tested on box, and compare your output with ours.
Every guide is run before it is published
- Run first. The complete code ran on a Snowflake Enterprise Edition sandbox, or in DuckDB for plain SQL, before the guide went up.
- Real output from two runs. You see the actual result of one run that catches an error we planted, and one that comes back clean.
- Sample data only. Each guide names its dataset in its Tested on box, such as the TPC-H benchmark sample. No client data, no client names, and every result shown is what the code actually returned.
- Check it yourself. A “Tested on” box gives the platform, edition, date and dataset. The code downloads as a .sql or .py file, and each guide links to the vendor’s own documentation.
Native Data Quality
Data quality checks that run inside the platform
Move your data quality rules out of separate jobs and into the checks Snowflake runs itself.
No guides here yet. Each one is published only after its code has run.
Metadata Enrichment Engine
Profiling, classifying and describing columns
Profile, classify and describe every column, so you know what it means and how sensitive it is.
No guides here yet. Each one is published only after its code has run.
Need checks like these running across your own data?
A guide shows one problem solved on sample data. Tell us what you are working on, and we will tell you whether a guide is enough or a solution fits.