How to Fix Your Technical Debt Dilemma
summary
Claim: automated, rules-based refactoring can fix large-scale technical debt by applying tested OpenRewrite recipes across repositories. The demo runs recipes on a public tenant of 300 million lines and on smaller sets (146 repos), handles junit4→5 and chained migrations like migrate to Java 17 and Spring Boot 1.5→3, and integrates with CI, GitHub and on‑prem agents. Generative edits (Copilot/LLMs) help but are probabilistic (≈70% then ≈85% with prompt tuning), so rule-based recipes remain the safe, repeatable core.
AI-written summary · about
chapters (8)
- 0:00Introduction to technical debt
- 4:36Automated code refactoring solutions
- 8:07OpenRewrite refactoring engine
- 14:33Idiomatic code consistency
- 18:30Writing a refactoring recipe
- 26:20Advanced recipe use cases
- 31:42Generative AI in refactoring
- 37:43Q&A session
related talks
45mEnable Digital Sovereignty by Tech Stack Liquidity
59mModernize Software and Tackle Tech Debt at Big Bank Scale
01h 17mRefactoring the ROI of Software
54mResolving technical debts in software architecture
43mDeliver Fast with Confidence
15mMajor migrations made easy
45mTrusting AI to Modernize Software at Scale
23mRefactoring Go in Large Codebases