CLPS Incorporation Sets Industry Benchmark with Completion of AI-Driven Modernization of a 30-Year-Old Legacy Banking System

CLPS Incorporation (the "Company" or "CLPS") (Nasdaq: CLPS) today announced the successful completion of an artificial intelligence (AI)-driven modernization project for a 30‑year‑old mortgage loan system at a major international bank (the "Client"). This milestone project demonstrates the transformative potential of AI in modernizing complex legacy environments and establishes a replicable success model for the global fintech industry.Project Challenge: Overcoming a 30-Year "Code Black Hole"As business requirements evolved, the Client urgently needed a comprehensive upgrade of its core mortgage loan system that had been in operation for over three decades. The project faced severe challenges: aside from raw source code, there was virtually no design or requirement documentation. The original developers had long departed, leaving the system's operational logic largely unknown.Furthermore, the system architecture was extremely large and complex, encompassing:138 Visual Basic (VB) functional programs 248 Microsoft (MS) Access programs 315 reports and 114 letter templates (including mail‑merge Word templates) 303 batch jobs, 25 COBOL programs, and 841 stored procedures Over 700,000 lines of code.Faced with this documentation vacuum, a traditional modernization approach was estimated to require 80 developers and five years to complete. The Client sought a more efficient, technology-driven breakthrough to tackle this daunting technical debt.AI‑Driven Methodology: Reverse Engineering Code to DocumentationTo close the documentation gap, CLPS introduced "Know Your System" (KYS), a proprietary, end-to-end AI solution engineered to parse and reverse engineer legacy source code into comprehensive functional design documentation. Its core workflow includes: Automated Analysis: Leveraging Large Language Models (LLMs) to parse legacy source code and automatically generate functional descriptions. Expert Verification: CLPS subject matter experts verified the output. Initial AI accuracy reached approximately 80%‑90% and improved to 98% following iterative prompt optimization. Targeted Documentation: The outputs were segmented into technical specifications for developers and business logic documents for operational stakeholders. Automated QA: Based on the newly generated documentation, AI was utilized to automate the generation of System Integration Testing (SIT), User Acceptance Testing (UAT), and Regression Test Cases.Project Outcomes: Breakthroughs in Efficiency and QualityThe project was delivered by just over 20 developers in 16 months, compared with an initial estimate of 80 developers over approximately five years. This represented a substantial reduction in delivery resources and project duration, resulting in a significant cost saving.The old system was successfully re-platformed to a modern Java/React/PostgreSQL technology stack. The migration scale included: Online Functions: Converted 386 VB and MS Access programs into modern Java and React applications (totaling approximately 449,732 lines of code), enhancing scalability and maintainability. Batch Processes: Migrated 303 batch jobs, 315 reports, 114 letter templates, and 841 stored MS procedures into modern Java‑based processes, JasperReports, MS Word templates, and PostgreSQL stored procedures (totaling approximately 535,131 lines of code), ensuring streamlined operations and improved maintainability. Data Migration: Successfully rebuilt 1,124 database tables from MS SQL Server and VSAM into PostgreSQL database. Legacy system data was seamlessly migrated, ensuring smooth transition and full operational continuity in the new Java based system environment.The modernized system comprises approximately 984,863 lines of code. CLPS successfully migrated the legacy applications and database to a scalable, enterprise‑grade platform.Technical Milestone: AI‑Powered Code RecoveryA notable technical challenge arose when the source code for certain VB modules was found to be missing, leaving only compiled Dynamic-Link Library (DLL) files. By cross-referencing code from systems with similar functionality and leveraging AI's deep understanding of the established business logic, the CLPS team successfully inferred and regenerated the missing functional logic. This edge case highlights AI's powerful capability in extreme‑complexity recovery scenarios.Security and Compliance: On-Premises Deployment for Banking ComplianceTo meet the Client's regulatory compliance and data security requirements, CLPS deployed a localized, on-premises AI solution within the Client's own environment. Under this deployment model, confidential client data was not transmitted to any third-party public cloud AI service. CLPS believes this approach addresses one of the principal obstacles to AI adoption among highly regulated financial institutions.This project saved the Client millions of dollars in direct costs while substantially reducing future maintenance overhead. More importantly, this success story sets a benchmark for the entire banking industry. Furthermore, CLPS's validated AI‑assisted development model is highly scalable across various industries relying on legacy systems built with traditional languages like VB and COBOL.Mr. Henry Li, President of CLPS, commented: "In the core strategic direction of legacy system modernization, we have developed a systematic COBOL‑to‑Java methodology. Since 2005, CLPS has focused on the maintenance, upgrade, and modernization of COBOL‑based core banking systems, accumulating extensive experience in complex migration projects. Integrating over two decades of banking domain expertise with cutting‑edge AI tools, we provide financial institutions with a highly scalable, secure path to reduce operating costs and accelerate digital transformation. We believe this shift toward AI-powered solutions will become a core competitive advantage for the Company, serving as a strong second growth curve for CLPS."

CLPS Incorporation Sets Industry Benchmark with Completion of AI-Driven Modernization of a 30-Year-Old Legacy Banking System

他社のニュース