Flezi EMT – AI-Enabled Legacy Modernization

Accelerate mainframe & midrange modernization with Flezi EMT the Unified AI-Enabled Solution that seamlessly integrates LLM to support 4R-AI-Enabled approaches to modernize legacy applications to open, no-vendor lock, and cloud native architecture.

FPT’s Enterprise Modernization Technology (EMT) Solution, powered by Azure OpenAI, helps enterprises modernize legacy mainframe systems such as COBOL, JCL, PL/I, RPG, and Assembler into modern, cloud-native applications written in Java, Shell, and Python.

By integrating Generative AI and Microsoft Azure capabilities, EMT enables organizations to analyze, document, refactor, and migrate legacy applications with up to 40-50% higher developer productivity, 40-50% lower AI infrastructure cost, and 30-45% reduced delivery risk compared to traditional approaches

Customer Pain Points Addressed

• High maintenance costs and dependency on legacy skillsets – outdated systems require rare expertise and high support effort.
• Incomplete documentation and knowledge loss – decades-old systems lack updated references, increasing modernization risk.
• Complex interdependencies between programs, data, and batch jobs – manual analysis slows down modernization and increases error rates.
• High risk and uncertainty in large-scale migration – traditional methods lack automation, visibility, and consistency for cloud readiness.

How FPT’s EMT Solution Addresses These Challenges

• EMT X-ray – Assessment & reverse engineering: builds asset inventory and migration strategy, and generates basic/detailed design from existing source code and documentation.
• EMT Bridge – Application & data migration: migrates COBOL/JCL/RPG to modern Java/ReactJS architectures and SAM/VSAM/DB2/IMS data to PostgreSQL, replacing IBM/Fujitsu/AS400 utilities.
• EMT Forge – Reverse & forward engineering: restructures business processes into M*-service architecture and builds new applications from AI-assisted specification and design.
• EMT vAlid – AI-powered testing: runs pattern, unit/integration, and comparison tests to validate legacy-to-modern migration outputs.
• EMT Run – DevOps & operations: builds, deploys, and scales cloud-native infrastructure (Azure/AWS/GCP) with continuous monitoring and failure prediction.
• EMT Shield – AI security gateway: masks PII and controls cost for AI/LLM calls, and scans code for security and secrets issues (SAST/SCA), independent of the AI engine.
• EMT BrAIn – The AI core connecting all EMT modules through Mentor, Co-pilot, and Worker modes, providing shared agents, an MCP server, and APIs across the modernization journey.

Outcome

• 40–50% higher Developer Productivity through AI-augmented engineering across the full modernization lifecycle.
• 30–45% lower Delivery Risk via deterministic code mapping and continuous AI-assisted validation.
• 40–50% lower AI Infrastructure Cost through governed, cost-optimized use of AI/LLM services.
• 90% of tasks fully automated with zero SME re-work needed across assessment, migration, and validation.
• ~40% total program effort reduction following FPT's unified 4R approach (Rewrite, Rebuild, Replace, Retain) with continuous governance from EMT BrAIn.