Caching, harnesses, and advisors: Building on Claude at GitHub scale

ℹ️ EPISODE INFO

Released

May 6, 2026

Duration

26m

Channel

Claude

Watch on YouTube

▶ Open ↗

📝 EPISODE DESCRIPTION

GitHub's Copilot team ships Claude to millions of developers across chat, CLI, coding agent, and code review, and has become one of the most demanding users of the Claude Platform. GitHub CPO Mario Rodriguez and Anthropic's Brad Abrams break down how the team pushes quality up and costs down at scale, from caching and evaluation to the new Advisor strategy. Walk away with patterns you can apply to your own Claude-powered product.

🎤 SPEAKERS

• Brad Abrams

  host

  Anthropic team member and on-stage presenter discussing Claude models, evaluation feedback, and GitHub Copilot integrations.

• Mario Rodriguez

  guest

  Chief Product Officer of GitHub speaking about building GitHub Copilot features at scale using Claude (caching, harnesses, and advisor models).

🧾 EPISODE SUMMARY

In this episode of Claude, featuring Brad Abrams and Mario Rodriguez, Caching, harnesses, and advisors: Building on Claude at GitHub scale explores gitHub’s Claude scaling playbook: caching, routing, and evaluation loops GitHub treats prompt caching as the primary cost-and-latency lever, targeting ~94–96% cache hit rates and considering ~70% a likely bug or prompt/tooling regression.

🔗 RELATED EPISODES

The CLAUDE.md file

MCP in Claude Code

What's new in Claude Code

Building with Claude on Google Cloud

The thinking lever

Building AI-native: Inside the stacks powering Cognition, Gamma, and Harvey