How ERGO Hestia reduced time-to-market with Lakebase and Mosaic AI Model Serving

ERGO Hestia, a leading Polish insurance company, redesigned its real-time pricing platform using Databricks Lakebase and Mosaic AI Model Serving to reduce model deployment time and operational complexity.

• •The previous multi-hop architecture routing data through external PostgreSQL and adapter layers created performance bottlenecks, latency spikes of 10-20x, and governance fragmentation across systems.
• •Lakebase provides a relational transactional layer on Delta tables with automatic synchronization via Sync Tables, eliminating manual extraction jobs and creating a single source of truth for pricing data.
• •Model Serving Endpoints expose models directly to applications, removing intermediate adapter layers and reducing request latency to milliseconds while consolidating query execution into a single managed layer.
• •The team employed incremental migration: Weeks 1-3 PoC achieved 20ms latency with <5% CPU utilization at 40 requests/second; Weeks 3-6 expanded to production endpoints; P