Contact
Home Cases Insurance StratusNex

Insurance

StratusNex

How we leveraged AI to update client’s old Java apps, and therefore reduced engineers’ manual effort by about 70%.

cover development

Technologies

Apache Struts, GitHub Copilot

Industry

Insurance

Engagement Model

Dedicated Team

Platform

Web

Team

NDA

Timeline

NDA

Client & Project Overview

The client managed several legacy Java applications built on Apache Struts 1.x. Several apps have been in production for over a decade. 

These Java apps used outdated Struts tags in JSP pages for their web interfaces. That dependence opened security risks and caused compatibility problems.

The client wanted to upgrade the Apache Struts library version to Struts 7.0.0 and also migrate the applications to the latest Java runtime. The move demanded many hours of hands-on refactoring and testing and was a big challenge.

They reached out with two requests:
  • Leverage AI tools to automate the migration process and avoid the slow, manual upgrades that take too long.
  • Include deep AI expertise in their workflow and keep their development processes up-to-date and ready for future changes.

Objectives & Strategic Approach

The project faced no major technical problems, but the client’s goals required creative solutions. The goals were ambitious, so we built a plan that balanced quality and speed with knowledge transfer.

We had to:
  • Show how AI can speed up the migration  – tags in Java Server Pages (JSP) code, Action classes.
  • Provide hands-on training in AI tools to help the client’s internal teams become more skilled and independent.

JetRuby proposed a Proof of Concept (PoC) that measured GitHub Copilot’s ability to automate and update code effectively. 

This plan fit the client’s Azure + GitHub stack perfectly.

Our Work: AI-Driven Proof of Concept

Backed by in-depth AI and PoC expertise, our team
  • Set up a development environment for a legacy application to match the client’s system.
  • Created a proof of concept in two weeks.
  • Automated the replacement of Struts tags in JSP pages using GitHub Copilot.
  • Trained GitHub Copilot to recognize old Struts 1.x tags and action classes, and update them to Struts 7.0.0 versions.
  • Improved the JSP page update process by creating smart prompts for Copilot, reducing the need for manual code rewriting.
We highlighted how Copilot helps simplify development tasks and speed up project growth
  • Demonstrated the Proof of Concept (PoC) workflow.
  • Showed how Copilot can handle repetitive tag migrations.
  • Provided steps for scaling the solution to other client's applications.
  • Shared ideas for using Copilot in code reviews and boilerplate code generation.

While working with the client, we demonstrated the PoC features and showed how Copilot handled repetitive tag updates. We explained how it helped with tasks like code reviews and standard code creation.

Our team wired Copilot to Azure, and suggested tools like Replit and Cursor for rapid prototyping in future projects.

We modernized and leveraged AI tools to update the client’s old systems while letting their team keep innovating.

Results & Long-Term Impact

The Proof of Concept provided fast and flexible results.

We upgraded an old application to Struts 7.0.0, showing that our AI update approach works. 

The client’s team started using GitHub Copilot to automate upgrades for other applications, reducing manual effort by about 70%. They also began using Copilot to accelerate other projects, from code reviews to feature development.

JetRuby’s expertise helped the client introduce AI to their software development process, allowing them to keep up with industry trends.

Why Choose JetRuby?

Our Expertise in Proof of Concept and AI Development

JetRuby quickly tests new ideas and ensures they work with your existing systems, like Azure and GitHub. We customize tools like GitHub Copilot to help solve specific problems, including updating old code.

We also help clients harness AI’s potential in real-world use. This support makes it easier for teams new to AI to understand and leverage AI-driven tools.

We focus on providing tailored AI solutions for businesses that want to improve workflows, lower costs, and grow effectively.

Whether you need a PoC for old systems or a full AI rollout, JetRuby delivers precise and measurable results.

Share
Link copied!

Follow us

behance Vector development

Follow us

dribble word development

Frequently Asked Questions

How to choose a system modernization company for legacy applications?

Choosing the right partner is a high-stakes decision that requires a multi-dimensional assessment of technical capabilities and strategic alignment. Key criteria include:
  • Technical Auditing. The partner must perform a comprehensive audit to map your current architecture and identify hidden bottlenecks before proposing changes.
  • Proven Track Record. Look for case studies demonstrating the successful handling of complex, large-scale projects to mitigate high-stakes risks.
  • Security & Compliance. Ensure they strictly adhere to global standards such as GDPR, HIPAA, and PCI-DSS, and implement secure coding practices.
  • Cloud-Readiness. A strategic partner should have a transparent roadmap for migrating applications to public, private, or hybrid clouds while managing data latency.

AI-augmented engineering reduces technical debt by leveraging AI tools to detect and eliminate security vulnerabilities with minimal human intervention. These tools can also prioritize debt by identifying "hotspots" where code changes most frequently, ensuring that modernization efforts deliver the highest impact on business velocity. Additionally, Large Language Models reconstruct lost architectural logic by auto-documenting legacy methods and summarizing complex dependencies.

GitHub Copilot customization for enterprises includes training custom models on private codebases to ensure suggestions align with proprietary internal libraries. Management can further refine output by adding natural language "Organization Instructions" to enforce global coding standards or repository-specific requirements.

AI coding assistants enable developers to complete tasks up to 55% faster while increasing pull request merge rates by approximately 15%. Beyond quantitative productivity, these tools improve the developer experience by reducing mental strain and helping users stay in a productive "flow state".

Contact us

By submitting request you agree to our Privacy Policy

By submitting request you agree to our Privacy Policy

Contact us

By submitting request you agree to our Privacy Policy