Educational platform for learning AI agent orchestration

Live Site: https://fbratten.github.io/8me-showcase Code Repository: https://github.com/fbratten/8me

8me teaches AI agent orchestration from fundamentals to production systems. Starting with simple loops and progressing to sophisticated multi-agent patterns.

This repository contains educational content with links to the code repository:

┌─────────────────────────────────────────────────────────────┐
│                   Bidirectional Linking                     │
├─────────────────────────────────────────────────────────────┤
│                                                             │
│   8me-showcase (this repo)      8me (code repo)            │
│   ┌─────────────────────┐      ┌─────────────────────┐     │
│   │ Tutorials           │◄────►│ src/tier0-hello...  │     │
│   │ Labs                │      │ src/tier1-ralph...  │     │
│   │ Concept docs        │      │ src/tier2-ralph...  │     │
│   │ Architecture guides │      │ src/tier3-mcp-...   │     │
│   └─────────────────────┘      └─────────────────────┘     │
│           │                              │                  │
│           │   "Get the code" ──────────► │                  │
│           │ ◄──────── "Learn more"       │                  │
│                                                             │
└─────────────────────────────────────────────────────────────┘

1. Brand new to loops? → Start with Your First Loop
2. Want working code? → Visit 8me repository
3. Learning patterns? → Read Orchestration Concepts
4. Hands-on practice? → Try Lab 01

• Lab 01: Your First Loop
• Lab 02: External State
• Lab 03: Simple Verification

• Lab 04: JSON Task Management
• Lab 05: Tool Calling Patterns
• Lab 06: Circuit Breakers
• Lab 07: Confidence-Based Retry

• Lab 08: Building an Orchestrator
• Lab 09: Multi-Agent Patterns
• Lab 10: Gating and Drift Prevention
• Lab 11: Self-Play Oscillation
• Lab 12: Memory Integration

• Lab 13: MCP Server Deployment
• Lab 14: Claude Code Skill Development
• Lab 15: End-to-End Workflow

"Loop persistently until the job is done."

The "Ralph Wiggum" methodology teaches that success comes from:

• Persistence: Keep trying until verified complete
• External verification: Don't trust AI self-assessment alone
• Safety limits: Circuit breakers prevent infinite loops

For the AI nerds: This loop-until-done pattern was the common approach before sophisticated orchestration frameworks. Sometimes the old ways persist because they work.

Note: Labs contain pedagogical variations with richer abstractions (interfaces, hooks, builders) optimized for teaching concepts layer-by-layer. The 8me source repo contains the "mundane" working implementations optimized for actual use.

# Clone this showcase repo
git clone https://github.com/fbratten/8me-showcase.git
cd 8me-showcase

# Serve locally (requires Jekyll)
bundle install
bundle exec jekyll serve

# View at http://localhost:4000

This is a personal educational project. See CONTRIBUTING.md for details.

MIT License - See LICENSE file for details.

Part of the 8me autonomous loop toolkit & Adaptivearts.ai™ initiative