Localhost

Local AI Homelab

Windows-native ROG Ally edge AI node. Browser talks to local Ollama through an SSH tunnel over Tailscale.

Run order

On the ROG Ally, open PowerShell as Administrator.
Scaffold the 4TB vault: ./scripts/scaffold-vault.ps1 -Path D:\ai-vault
Install/configure Ollama and pull qwen: ./scripts/setup-ollama.ps1
On the main laptop, open the tunnel: ./scripts/start-tunnel.ps1 -RogHost <ROG_TAILSCALE_IP> -User <ROG_WINDOWS_USER>
Start the UI: cd web-ui docker compose up -d
Open http://localhost:3000 or https://homelab.himanshulade.com.

Chain

browser -> localhost:11434 -> SSH tunnel -> Tailscale -> ROG Ally -> Ollama

Notes

ROG Ally stays Windows-only. No WSL dependency.
Ollama models are stored under D:\ai-vault\models\ollama.
Allow inbound TCP 11434 in Windows Firewall on the ROG Ally.
nginx config expects Cloudflare proxy with Full (strict) TLS.

Let's make the next system less fragile.

Open to software engineering roles across full-stack systems, platform and reliability work, workflow automation, and applied AI. I value teams where I can keep learning while contributing to real systems and clear delivery outcomes.

Also open to freelance or contract work across full-stack builds, practical automation, technical SEO, and cloud delivery.

Repo notes

What the build looked like.