Video by FINOS via YouTube
Liborio Ciccarello (Technology Director, Full Stack Engineer & DevOps Lead at Morgan Stanley) demonstrates how large financial institutions can solve the exponential complexity of enterprise code refactoring. Liborio shares Morgan Stanley’s real-world modernization journey using OpenRewrite and Moderne, leveraging Lossless Semantic Trees (LST) to execute deterministic, scale-wide migrations—upgrading Java and Spring Boot codebases across thousands of repositories while enforcing SDLC compliance and eliminating tech debt without token limits or build breaks.
🗽 Catch Us in New York! Ready to eliminate enterprise technical debt, automate zero-day CVE remediations, and scale deterministic software hygiene?
Join global technology leaders at OSFF New York on November 4–5, 2026.
🎟️ Register Now: https://hubs.ly/Q04n_bZL0
🔥 20% OFF DISCOUNT CODE: 26YTOSFFNY20C
🕒 Timestamps:
0:00 Introduction: Enterprise Scale & Code Modernization Challenges
0:40 The Manual Scaling Curve: Repository Explosion & Tech Debt
1:35 The Pitfalls of Manual Hygiene: Resource Constraints & Priority Collisions
2:12 OpenRewrite vs. Moderne: Lossless Semantic Trees (LST) Explained
3:22 Real-World Case Study: Java 11 & Spring Boot 2 to Java 21 & Spring Boot 3
4:30 Beyond Upgrades: Code Health, Security Best Practices & Cleaning Dependencies
5:20 Writing Custom Recipes: Tooling Onboarding & CI/CD Pipeline Transformation
6:40 Key Takeaways: Deterministic Refactoring & AI Co-pilot Execution
7:40 FINOS Call to Action: Building Enterprise Financial Recipes Together
8:00 Q&A: Scaling Across 50,000 Repositories & CVE Remediation
📊 The Problem: The Exponential Enterprise Migration Wall
Upgrading a few repositories from Java 11 to Java 21 or Spring Boot 2 to 3 is straightforward when done manually. However, at enterprise scale across tens of thousands of repositories (e.g., Morgan Stanley’s 50,000+ repos), manual hygiene collapses under resource constraints, delivery collisions, partial adoption, and accumulating tech debt. While generative AI LLM agents can help, relying purely on probabilistic LLM code generation risks build breakage, inconsistent formatting, high token costs, and context window limits.
🏗️ The Solution: Lossless Semantic Trees (LST) & Scaled Refactoring Recipes
Liborio Ciccarello outlines how OpenRewrite and Moderne deliver deterministic code transformations: Lossless Semantic Trees (LST): Parses codebases into exact semantic trees that preserve every comment, formatting detail, and whitespace character during refactoring. OpenRewrite vs. Moderne: OpenRewrite provides open-source, single-repo Gradle/Maven refactoring recipes. Moderne scales this architecture across multi-repo, multi-recipe enterprise fleets via LST persistence without context limits. Custom Enterprise Recipes: Enables DevOps leads to write custom, testable recipes to automate static analysis tool onboarding and transform weekly CI/CD pipeline builds across the entire bank. AI Integration (Prethink): Pairs LLMs for recipe authoring while relying on Moderne for deterministic execution, ensuring 100% predictable code changes across thousands of repositories in seconds.
⚙️ Why This Matters for Financial EngineeringInstant Zero-Day CVE Remediation:
Allows security and platform engineering teams to push declarative recipes that fix critical vulnerability zero-days across thousands of codebases concurrently. Frees Up Developer Capacity: Automates repetitive framework migrations, dependency cleanups, and static analysis fixes, allowing engineering teams to reallocate focus toward core business deliverables.
🌐 More about FINOS: https://www.finos.org/
📧 Join our newsletter: https://www.finos.org/sign-up
🎙️ Listen to our Open Source in Finance Podcast: https://www.youtube.com/@FINOS/podcasts
LinkedIn: https://www.linkedin.com/company/finosfoundation
#FINOS #OSFFNewYork #MorganStanley #OpenRewrite #Moderne #DevOps #Java #SpringBoot #SoftwareModernization #FinTech #DevSecOps