Field intelligence for AI-first professionalsVol. II · Nº 56 · Saturday, August 15, 2026
← All tracks
Track 37Pro Track~120 min total

Build AI Agent Teams with Claude Code

Master multi-agent orchestration with Claude Code. Learn subagents for parallel tasks, Agent Teams for coordinated workflows, the Claude Agent SDK for programmatic control, and advanced patterns like autonomous /loop runs and git worktree isolation.

Prerequisite:

§  The modules
  1. 01When One Agent Isn't EnoughThe scaling problem with single-agent workflows. When to split work across agents. Subagents vs Agent Teams - the 60-second decision framework.Account
  2. 02Building Your First SubagentThe .claude/agents/ directory, YAML frontmatter, tool restrictions, foreground vs background, and the built-in subagent types.Account
  3. 03Subagent Patterns and Best PracticesParallel research, code review pipelines, context isolation, memory persistence, model selection, and choosing the right orchestrator model.Pro
  4. 04Agent Teams Setup and CoordinationEnable Agent Teams, create teammates, shared task lists, inter-agent messaging, display modes, and hands-on team setup.Pro
  5. 05Building Real Workflows with Agent TeamsContent production pipeline, development pipeline, hybrid patterns with autonomous loops, task dependencies, and quality gates.Pro
  6. 06The Claude Agent SDKPython and TypeScript setup, the query() function, programmatic subagents, building custom pipelines, and when to use SDK vs CLI.Pro
  7. 07Advanced PatternsGit worktree parallel sessions, autonomous loops (/loop and the Ralph technique), hooks for coordination, and combining patterns for complex workflows.Pro
  8. 08Using Pre-Built Team TemplatesImport LLM Match Maker team templates, customize personas for your workflow, set up as Agent Teams, and connect to Paperclip for persistent orchestration.Pro
Start module 01 →Finish all six for the Pro Track