

Problem Solved!: OpenClaw Workshop Part 2
🦞 OpenClaw Part 1 — Architecture & The Harness Part 1 of 5 · Organized AI Workshop Series
What if your AI assistant lived on your own machine, talked to you through WhatsApp, and got smarter every conversation — without a single subscription?
OpenClaw is a local-first AI gateway that routes messages from 20+ channels to an AI agent running on your hardware. In Part 1 we’ll pull back the curtain on how it’s built, why it’s built that way, and how you can start running your own. Whether you’re here out of curiosity or ready to fork the repo, there’s a layer for you.
🗓 Agenda
First Half — Understand the System
• Welcome — What is a personal AI gateway, why local-first matters, and the mental model that ties everything together • The Architecture — Plain-English walkthrough of how a WhatsApp message travels from your phone through Cloudflare, into a Durable Object, and back to you in seconds — and why each layer is there • The Harness — How the system observes itself: every message is logged with a difficulty tier, token count, and tool calls. Use !good / !bad / !pin to teach it what works • Own Your Data — Your state lives in R2. When you’re ready to self-host, your full history exports with you
~5 min break
Second Half — Build It
• Get Connected — Join the workshop WhatsApp gateway, send your first message, watch it route live • See the Logs — Peek inside prompt_logs and see what the system captured about your turn • Open Floor — Live debugging, architecture questions, help getting set up
What’s Actually Running
Your message doesn’t go straight to an AI. Here’s the real path:
You (WhatsApp) ↓ Cloudflare Worker ← receives from Meta Cloud API ↓ CF Queue ← guarantees delivery, handles retries ↓ Durable Object (you) ← your SOUL.md, memory, session state ↓ OpenClaw on local machine ← the AI gateway, Tailscale-connected ↓ prompt_logs ← every turn logged, tiered, observable
If that’s too much detail right now — no worries. By the end of the session it’ll make complete sense. If you already know what Durable Objects are, you’ll appreciate why this pattern was chosen over a simpler queue-only approach.
What You’ll Walk Away With
✅ A clear mental model of how personal AI gateways actually work ✅ Understanding of why local-first AI is fundamentally different from ChatGPT or Siri ✅ Your first hands-on experience sending messages through a real production gateway ✅ A foundation for everything in Parts 2–5
Who This Is For
Curious non-technical folks who want to understand what’s under the hood — and builders who want to see a production-grade architecture before they start wiring things up. Part 1 is designed so both leave with something useful.
Three Learning Modes
📖 Docs Only — No install needed. Diagrams and concept explainers. ⭐ Diagrams + Docs — Mermaid diagrams and annotated examples. Recommended starting point. 💻 Full Hands-On — Live Gateway + channel exercises. OpenClaw required.
The Series
Each session builds on the last. Start here, leave with a foundation.
Full guide → openclaw-education.pages.dev
Your Presenter
Jordan Hill is the founder of Organized AI and Blue Highlighted Text, an AI automation and consulting agency based in Austin, TX. He builds agentic infrastructure for real businesses — from conversion tracking pipelines to self-hosted AI assistants — and created this workshop series to make personal AI architecture accessible to anyone willing to get their hands dirty. This workshop runs on the infrastructure he built.
Organized AI · The Lobster Way 🦞🥷🏾