Session

Beyond Vibe Coding: An AI Builder’s Guide to Migrating 10k+ React Components

with Shubham Chaurasia

About This Session

Ads Manager powers 84% of Meta's Ads Revenue, but its decade-old codebase was stuck on React class components — blocking modern patterns like hooks, Relay, and React Compiler. Manual migration of 10k+ components class to functional component was impractical. In this talk, I'll show how we built an AI codemod pipeline using LLMs that achieved a 5x speedup, migrating 3k+ components with a 99.6% success rate. I'll cover the tradeoffs between AST and AI codemods, how to classify components by complexity for targeted prompts, and the grading/eval strategies that made non-deterministic AI agents production-ready.

Topics

AI Coding Assistants

← Back to Schedule

Speaker

Shubham Chaurasia

Sr. Software Engineer · Meta

Sr. Software Engineer at Meta

Read bio Hide bio

As a Senior Software Engineer at Meta, I thrive at the intersection of complex systems, web performance, and developer experience. My career is defined by a commitment to engineering excellence—from driving large-scale architectural transformations to leveraging emerging AI technologies to solve the industry's toughest challenges. Currently, I contribute to the architecture of Meta's Ads Manager, the world's largest React application. My work centers on ensuring top-tier performance and reliability for this mission-critical platform, employing rigorous observability practices to minimize browser crashes and optimize user experience at scale. My recent focus centers on "AI-assisted engineering." By building production-grade AI codemod pipelines, I have demonstrated how non-deterministic agents can be constrained to perform reliable, deterministic transformations—migrating over 3,000 components with a 99.6% success rate, significantly reducing technical debt while maintaining velocity for over 1,200 active developers. With a background spanning engineering leadership at Groupon to infrastructure work at Meta, I specialize in re-architecting consumer-facing applications and fostering high-performance engineering cultures. Whether it's diagnosing complex performance bottlenecks or designing scalable AI-driven development workflows, I am driven by the goal of making engineering more efficient, reliable, and developer-friendly.