FINOPS / 04
SQL pre-flight for Snowflake cost
Profiling warehouse spend, then shipping a Streamlit-in-Snowflake tool that flags expensive query patterns before they execute.
pre-executioncost signal
anti-patterndetection
self-servefor developers
01 / THE CHALLENGEMake the repeatable path
Make the repeatable path
the default path.
Warehouse spend was rising from patterns no dashboard caught in time: CTAS recreation loops, result-cache misses and SQL anti-patterns that only surfaced on the monthly bill.
02 / THE ARCHITECTURESystems that stay
Systems that stay
legible under load.
Spend was profiled to isolate the recurring cost drivers. Those patterns became rules inside a Streamlit-in-Snowflake app, so developers can paste or route a query through a pre-flight check that explains why it is expensive and how to rewrite it — before it runs.
Cost control moves left — developers see the expensive pattern before the warehouse bills for it.