CCA-F: Claude Certified Architect — Foundations

Question

You are creating developer-productivity tools with the Claude Agent SDK. The agent assists engineers in exploring unfamiliar codebases, understanding legacy systems, generating boilerplate code, and automating repetitive work. It uses the built-in tools (Read, Write, Bash, Grep, Glob) and integrates with Model Context Protocol (MCP) servers.

Your agent has spent 25 minutes examining a game engine’s rendering subsystem—reading shader code, buffer management, and frame-synchronization logic. An engineer now asks it to understand how the physics engine integrates with rendering for collision debug overlays. You notice that recent responses refer to “typical rendering patterns” rather than to the specific VulkanPipeline and FrameGraph classes it found earlier.

What is the most effective approach?

Explanation

A focused summary transfers the rendering facts needed to investigate the collision-overlay integration, while a sub-agent performs the new physics exploration in its own context window. This limits additional exploration noise in the main context and retains the specific rendering details required to connect the two subsystems. Anthropic documents subagents as a way to preserve context by keeping exploration separate and returning only a summary.

Learn more

Create custom subagents — Claude Code Docs

CCA-F: Claude Certified Architect — Foundations

AstroTutor

Notes

Readiness

QuestionQ83

Community Discussion