Unified Governance for Enterprise Analytics and AI
Overview
| Experience | In Person |
|---|---|
| Track | Governance & Security |
| Industry | Enterprise Technology |
| Technologies | AI/BI, Databricks SQL, Unity Catalog |
| Skill Level | Intermediate |
We implemented a unified governance architecture for enterprise analytics leveraging Databricks Unity Catalog as the authoritative control plane. The platform designed with a configuration-driven, dynamic attribute-based access control (ABAC) model with fully enforced row-level security (RLS) and column-level security (CLS) ensuring precise, policy-driven data access across the organization — including hierarchical visibility for cost owners and business leaders aligned to their reporting structures. This design prevents governance drift, reduces operational complexity, centralizes policy management, and enables secure self-service analytics across domains. Governance policies extend natively to AI/BI dashboards and Databricks Genie, ensuring consistent, compliant data access without duplicating or redefining security frameworks.
Impact
- Eliminated decentralized access control models across 40+ dashboards.
- Enabled secure self-service analytics for 300+ users across 20+ teams.
Session Speakers
Anand Tirthgirikar
/Staff Engineer
Inmobi