World Congress 2025 Aug 20, 2025 Session details

Migrating from COBOL with AI: A Moonshot Demo

Julia Kordick

Still terrified of touching massive legacy mainframes? Discover how a fault-tolerant, multi-agent AI framework systematically reverse-engineers and translates complex COBOL programs into modern Java.

Pause
Mute Enter Fullscreen
#1 about 2 min

Multi-agent AI for migrating legacy COBOL to Java

An open-source project uses multi-agent AI to help financial institutions modernize legacy mainframe codebases.

#2 about 1 min

Accessing the repository and collaborating on COBOL problems

Developers can access the project resources on GitHub and collaborate on solving complex legacy codebase problems.

#3 about 2 min

Setting up the environment and executing the migration tool

The required setup involves configuring an AI endpoint and using a provided dev container for seamless execution.

#4 about 1 min

Understanding the multi-agent flow and COBOL code analysis

The initial analyzer agent performs reverse engineering and preparation steps on legacy source files to begin migration.

#5 about 2 min

Parsing dependencies and complex COBOL call chains

The dependency analyzer agent maps unconditional jumps and intricate call chains between legacy files.

#6 about 1 min

Orchestrating AI agents with the Semantic Kernel framework

The project utilizes Semantic Kernel in C# instead of Autogen to manage complex agent logs and orchestration effectively.

#7 about 1 min

Storing intermediate results for fault-tolerant AI migrations

The system stores intermediate state in markdown files to gracefully handle errors during long-running code conversions.

#8 about 1 min

Extending the framework beyond COBOL to Java migrations

The flexible framework allows developers to target different source and destination languages by adjusting agent personas.

#9 about 2 min

Reviewing generated Java code and migration documentation

An inspection of the output directory reveals the newly generated Java code, comprehensive migration reports, and architecture diagrams.

Matching moments

6:35 min

Modernizing legacy COBOL mainframe systems using AI agents

Chris Heilmann +2 · LIVE

3:23 min

Migrating legacy COBOL systems using generative AI tools

Chris Heilmann +9 · LIVE

4:40 min

Analyzing COBOL mainframe migration challenges with AI

Michael Friedrich Michael Friedrich · WWC Europe 2026

2:46 min

Deploying AI agents for enterprise legacy code modernization

Neel Sundaresan Neel Sundaresan +1 · WWC Europe 2026

2:15 min

Leveraging artificial intelligence to accelerate system migration

Annette Schelten Annette Schelten +1 · WWC 2025

3:03 min

Evaluating AI agents for iterative legacy code modernization

Michael Friedrich Michael Friedrich · WWC Europe 2026

Upcoming sessions on this topic

Open session

World Congress 2026 North America

DeepAgents: Build Multi-Agent AI Systems That Actually Work

Anjana Umapathy, Apoorva Jaiswal, Anagha Rumade

Anjana Umapathy
Apoorva Jaiswal
Anagha Rumade
Open session

World Congress 2026 North America

Beyond Vibe Coding: An AI Builder’s Guide to Migrating 10k+ React Components

Shubham Chaurasia

Senior Software Engineer @ Meta

Shubham Chaurasia
Open session

World Congress 2026 North America

Look What Java Can Do Now: Live-Coding a GenAI MCP Server with the JAQ Stack

Suren Konathala

Senior Engineering Leader, Marketing & AdTech, AI GTM & Digital Experience Platforms, Java x AI

Suren Konathala
Open session

World Congress 2026 North America

Closing the Visibility Gap: Lessons from Safety Critical Agentic Systems

Vivek Pandit

Principal Engineer at Cadence

Vivek Pandit
Open session

World Congress 2026 North America

AI-Powered Incident Triage: How We Built GenAI Agents with MCPs to Automate On-Call Workflows

Prakshal Doshi

Site Reliability Engineer

Prakshal Doshi
Open session

World Congress 2026 North America

When Agents Became Users: Rearchitecting Identity and Permissions for AI at Scale

Dor Cohen, Yoav Gal

Dor Cohen
Yoav Gal