Every enterprise sits on a "Cave of Wonders".A massive codebase packed with decades of invaluable business logic, hardened edge cases, and customer data. But it's usally locked behind outdated languages, undocumented dependencies, and crushing technical debt. For years, engineers have been forced to act as poor woodcutters, endlessly chopping away at Jira tickets just to keep these systems alive. That era is over. It is time to speak the new magic words: Agentic AI In this talk, we will explore how the shift from writing code to orchestrating AI agents is fundamentally changing software modernization. Using the metaphor of Ali Baba and the Forty Thieves, we will break down how to deploy multi-agent systems to safely raid the legacy cave. You will learn how to command a crew of specialized AI agents to autonomously map architecture, generate missing tests, and translate outdated monoliths into modern micro-services. Abdel will walk you through a real-life example of a team of six engineers modernized a Ruby on Rails application to three programming languages, compared performances, implemented a cloud native architecture and demoed a working prototype to leadership in five days. Using AI Agents to research, code and test along the way. Key Takeaways: - Assembling the Crew: How to design a multi-agent architecture where specialized agents handle translation, testing, and documentation. Orchestrated by human engineers. - Avoiding Cassim’s Trap: Why code is not the only thing that matters in application lifecycle and how the toil of dependencies can break or make a migration project. - The Morgiana Protocol: How the role of the software engineer is evolving from coder to orchestrator, relying on human-in-the-loop workflows and strict testing to catch hallucinations before they hit production.

Session Speakers