Making Code Automations at Scale: Leveraging AI for Code Migrations
summary
Claims automated codemods save 100s of engineering hours by generating migration scripts via LLMs that run from a CLI or a VS Code extension; the public demo at codemod.com/studio accepts before/after examples and emits runnable JS codemod scripts, with a registry to share them. Current scope focuses on JavaScript migrations (React and dependency upgrades), cites MSW v1→v2 done in hours, and reports enterprise interest (Netlify, call.com); platform is live and free with iterative AI validation.
AI-written summary · about
chapters (10)
- 0:00Introduction to code migrations
- 2:40Current landscape of code upgrades
- 4:02Challenges in manual migrations
- 6:08AI's role in code automation
- 7:51What are code mods
- 10:38Benefits of code modifications
- 12:27CodeMod Studio demo
- 13:30CodeMod Registry explained
- 14:34Platform walkthrough and community
- 15:57Q&A on code migration process
related talks
21mAutomating Front-End Refactoring
27mHow Codemods Turned a Nightmare into a 3-Minute Migration
40mRefactoring large JavaScript codebases
34mCodemods in the Era of AI
28mAI Migration at SCALE: Make Complex Code Tasks 10X Faster (The Blueprint)
29mLarge Scale Migrations Using Recast and jscodeshift to Scale
28mCross-File Codemodding
47mComputer, refactor my code!