case study

January 14, 2025

Generative AI-Powered Migration: How Quantiphi Accelerated Actuarial and Risk Model Migration for a Large Insurer

Read how Quantiphi helped a leading insurer migrate thousands of interconnected actuarial and risk models from SAS to PySpark—all within a year. Facing a complex legacy system and a tight deadline, the insurer needed speed and precision. With its GenAI-powered accelerators, Quantiphi streamlined the migration, automating code conversion, validation, and debugging—ensuring a seamless transition with accuracy and efficiency.

About the Client

A global insurance provider with a diverse portfolio of financial products and services. The company relies on complex actuarial and risk models to support decision-making, regulatory compliance, and long-term financial planning.

The Problem: A Complex Legacy in a Race Against Time

The organization’s existing SAS system was vital to manage all its actuarial and risk models but no longer agile enough to meet modern demands. Migrating to PySpark was the clear solution, offering scalability and speed, but the journey wasn’t simple. They were up against:

  • Tight Timelines: The migration had to be completed within a year.
  • Complex Features: The SAS codebase contained a large number of intricate and interdependent features.
  • Data Gaps: The absence of a Pentaho dataset added another layer of difficulty, limiting model fine-tuning for accurate code translation.

The Solution: AI Accelerators Leading the Way

Quantiphi brought its generative AI-powered accelerator suite, Codeaira, to the forefront. Designed to streamline and automate complex migration tasks, Codeaira’s AI-powered developer assistant transformed what seemed impossible into a seamless process. Here's how it worked:

  • Code Generator: Automated the generation of PySpark code, eliminating manual re-writing.
  • CodeXcelerate: Enabled fast and reliable translation of SAS code to PySpark.
  • QodeQure: Detected and resolved bugs swiftly to ensure clean, error-free code.
  • QlosureEase: Created unit test cases, validating the functionality of the converted code.
  • Synthetic Data Generator: Generated test data for robust validation and debugging.

Quantiphi also leveraged AWS services, including Amazon EC2, Amazon S3, SageMaker, and more, to power this migration, ensuring a scalable and secure processing environment.

Technology Used

Amazon EC2

Amazon EC2

Amazon S3

Amazon S3

Application Load Balancer

Application Load Balancer

Amazon Route 53

Amazon Route 53

Amazon SageMaker

Amazon SageMaker

AWS Identity and Access Management

AWS Identity and Access Management

Elastic Container Registry

Elastic Container Registry

The Results: Precision, Speed, and Efficiency

With Quantiphi’s solution, the enterprise not only met its goals but exceeded expectations:

  • >90% Confidence Score for the converted PySpark codebase, ensuring reliability.
  • A notable boost in developer efficiency, reducing time spent on manual code validation.
  • 3x reduction in migration timelines.

A New Standard for Code Migration

This wasn’t just a migration; it was a transformation in how the organization approached code modernization. By leveraging the power of AI, they turned a daunting challenge into an opportunity to innovate.

This project showcases the potential of combining AI-driven accelerators with cloud capabilities to solve real-world challenges. Could your business be the next to benefit from Generative AI-powered modernization? Let’s start the conversation.

Start Your Next Gen AI Journey Today

Discover how Quantiphi’s AI-powered solutions can transform your business. Fill out the form, and we’ll help you explore tailored AI strategies to unlock new opportunities for growth.

Thank you for reaching out to us!

Our experts will be in touch with you shortly.

In the meantime, explore our insightful blogs and case studies.

Something went wrong!

Please try it again.

Share