On demand
Plan First, Ship Faster: How CodeRabbit Built Agent Orchestration on Claude
AI coding tools helped engineering teams ship faster, but faster doesn't mean aligned. CodeRabbit saw this firsthand across millions of code reviews: AI-generated code produces 1.7x more issues than human-written code, and most of those issues trace back to misalignment between what the developer meant and what the agent build to solve it, they built an orchestration layer on Claude that moves the hard decisions upstream, turning vague requirements into structured, reviewable plans before coding agents start executing.
Join CodeRabbit, VP of AI David Loker for a live look at how that system works and what builders can learn from the architecture.
Featured speakers

David Loker
VP of AI @ CodeRabbit

Brittney Tong
Growth GTM @ Anthropic

Ethan Dixon
Applied AI @ Anthropic
What you’ll learn
Why the biggest cost of AI-assisted development isn't tokens or model quality, but rework from misaligned intent
How CodeRabbit's planning agent decomposes issues into structured plans and validates them before handing off to coding agents
A practical framework for building orchestration patterns into your own agent workflows