How Agentic Coding Can Help You Migrate Java Applications Faster
summary
Shows aworked example of migrating Java 11→21 using an agentic workflow that loads skills on demand (Klein + Oracle Code Assist), discovers current JDK and 3rd‑party versions, generates file diffs to update deprecated APIs, runs Maven builds and replays failures until they pass, and can add integration tests to raise coverage. Claims faster, higher‑confidence upgrades by automating dependency discovery, contextual planning, and CI validation; includes token‑performance data (≈97% at 8–10k tokens, ≈37% at 500k–1M).
AI-written summary · about
chapters (11)
- 0:00Introduction to Java upgrades
- 1:00Why Java upgrades are critical
- 2:59Challenges in legacy upgrades
- 5:08AI-assisted evolution
- 9:31Framework for modernization
- 14:48The role of context engineering
- 19:05Anatomy of a Cline request
- 22:15Customizing agents with skills
- 25:59Executing the migration workflow
- 33:28Ensuring stability and success
- 35:59Q&A and final takeaways
related talks
59mTools & Tips to Modernize Your Java code with AI
53mUpgrade Your Java with AI: From Legacy Code to High-Performance Workloads
50mJava. Migrating to 11 in real app
47mJava. Migrating to 11 in real app
53mApproaching the JDK 25 LTS release through real-world migration
28mAutomated Java Upgrades: From Pain to Productivity in 30 Minutes
51mRefactor Your Code to Java 20
46mBeam Me Up, Java! A Journey through Java LTS Versions 11 to 21