— Infrastructure stack · what powers this
Every hub on this site runs on a dual-machine AI stack. Not slideware. Production.
Most consultants pitch from a laptop. I pitch from a dedicated AI runtime. A Mac Pro on the desk for long-running local agents and inference, a Windows orchestrator for deploy, and a cloud layer (Vercel · Firebase · GitHub Pages) for delivery. When you hire me, you hire this stack.
iMac Pro · AI Engine
LOCAL · MELBOURNE
Ollama · gemma3:4b
localhost:11434 · 3.3 GB · CPU inference
RUNNING
OC-1 · OpenClaw-1
Discovery + drafting · Telegram bot · 6 hr poll
READY
HC · Hermes
Monitor + verify · daily URL sweep · drift detection
READY
Claude Desktop
Conversation + planning surface
ACTIVE
Hardware
Mac Pro · Xeon W · 64 GB RAM
UP
Windows Workstation · Orchestrator
PRIMARY · DEPLOY
Claude Code CLI
Primary orchestrator · ships this site
ACTIVE
Gemini CLI · Codex · Copilot
Multi-agent AIR Team coordination
ACTIVE
Firebase CLI · Vercel CLI · gh
Deploy paths · 7 active Firebase projects
ACTIVE
Cloud delivery layer
Vercel · Firebase Hosting · GitHub Pages
CLOUD
Tailscale mesh
Windows ↔ iMac Pro · secure sync
MESH