Home
Notes from what I’m building.
By day I work on Android platform engineering at Pinterest — design systems, modularization, the stuff that scales to millions of users. By night I point that same instinct at my own homelab: self-hosted infrastructure, local LLMs, AI agent experiments, and the occasional VPN rabbit hole.
A few places to start:
- Building a Local LLM Server on Blackwell: Everything That Went Wrong
- Your VPN Isn’t as Invisible as You Think
- Twingate vs Tailscale: Which Should You Use for Your Home Lab?
Topics I write about: homelab, AI, OpenClaw.
Posts
-
Sandy: An AI Assistant That Can't Leave the Sandbox
-
Xid 79, Six Days Later: The Clock Lock Held, and the Motherboard Stays in the Box
-
The GPU Fell Off the Bus, Again (and Again): Chasing Xid 79 to a Silicon RMA
-
Xid 79: A Stress Test That Passed for the Wrong Reason, and a Reddit Tip That Might Fix It
-
The GPU Fell Off the Bus: A Postmortem on the Blackwell Box's First Hardware Failure
-
How to Create Atomic PRs Using AI Agents: A Spec-Driven Approach
-
Building a Local LLM Server on Blackwell: Everything That Went Wrong
-
TrueNAS to UNAS Pro: An Append-Only Backup Story
-
Twingate vs Tailscale: Which Should You Use for Your Home Lab?
-
Anthropic Changed the Rules: Revising My OpenClaw Split Deployment for Zero Cloud Spend
-
Your VPN Isn't as Invisible as You Think: How My Router Knew I Was Torrenting
-
Building a Split-Brain OpenClaw Deployment: Gateway + Inference Server Over Tailscale
-
OpenClaw Hardware Showdown: Mini PC vs. Gaming Rig — When a Discrete GPU Changes the Equation
-
Running a Full Homelab on a Mac Mini M4 — No Proxmox, No Rack, Just Docker
-
OpenClaw’s Brain Transplant: A Deep Dive into Open Source AI Agent Hardware and Hardened Deployment
-
OpenClaw Exploration Plan with Base Mac Mini M4
subscribe via RSS