CASE STUDY

Global Reinsurer: Catastrophe Modelling on Snowflake + Vertex AI

A global reinsurer rebuilt catastrophe modelling on Snowflake + Google Vertex AI. Model run-time reduced 14×; new exposure analysis turnaround days → hours.

May 9, 2026 · 2 min read · By admin

At a glance

Industry: BFSI
Topics: Data Warehousing · GCP · Predictive Analytics
Published: May 2026
Headline outcome: 14 Model run-time: 14× faster

Measured outcomes

What this programme actually delivered.

Model run-time: 14× faster

The challenge

A global reinsurer’s catastrophe modelling ran on legacy infrastructure. Each new exposure scenario took 3-7 days to model. Real-time pricing during hard markets was impossible.

What we built

Snowflake-based modelling platform + Vertex AI for ML-augmented scenario generation. Pre-computed exposure cubes. Underwriter-facing UI for scenario exploration.