Immuta abstracts policies across Snowflake Databricks
Zachary Friedman, associate director of product management at Immuta, on security in the modern data stack
The key point is that Snowflake and Databricks are fighting to own the same control point in the modern data stack, the place where data is stored, queried, governed, and increasingly used to build AI products. That is why each keeps expanding beyond its original core. Snowflake started as the easier warehouse for SQL analytics, while Databricks started with data engineering and ML, but both now sell warehousing, governance, and AI tooling into the same enterprise accounts. Immuta sits on top of that overlap, translating one access policy across both systems when large companies run both.
-
Snowflake and Databricks overlap at the buyer and workflow level. A large bank or pharma company may use Snowflake for analyst queries and Databricks for pipelines and model work, but the same platform, security, and data leaders still have to decide where policy is written and enforced.
-
Databricks has moved directly into Snowflake territory with Databricks SQL and Unity Catalog. Unity Catalog is not just admin plumbing, it is the permissions layer that controls who can see which tables, rows, and columns, which makes it central to both analytics and AI workloads.
-
This rivalry creates room for a neutral layer like Immuta. When a company uses Snowflake, Databricks, Redshift, and BigQuery at once, Immuta lets a security team write one business rule and push it into each platform instead of rebuilding the same rule in four different permission systems.
The market is heading toward more direct platform competition, not less. As governance becomes part of the core product in both Snowflake and Databricks, each will try to pull more workloads onto its own stack. That makes cross platform security and policy abstraction more valuable for enterprises that do not want one vendor to dictate every data workflow.