Laser and other tech projects,
always free for your use.
Every post has the details, process, and exact settings to duplicate it yourself. The site is live and maintained — and most of the work here, including this site itself, is done with LLMs.
Latest projects
Builds, guides, and experiments from the shop.
Pixel-Art Widget Showdown: Three Coding Agents, One Brief, Two Authors
I gave dsh, Kimi K3 and MiniMax M3 the same brief for a pixel-art LaserLloyd logo widget. dsh and Kimi wrote working widgets. MiniMax replied that it had built one, but the files it pointed to were Kimi's.
MiniMax M3 as an OpenClaw Coding Agent: Five Tasks Against Kimi K3 and dsh, and a Widget It Didn't Build
MiniMax M3 on OpenClaw's own agent loop passed five small coding tasks in 159 s, about half Kimi K3's time and a quarter of its list-price cost. The pixel-art widget I first credited to it turned out to be Kimi's.
Kimi K3 on Moonshot as a Coding Agent: Wiring It Into OpenClaw, Benchmarking It Against dsh, and Asking It for a Pixel-Art Logo Widget
Moonshot's Kimi K3 driving OpenClaw's native agent loop on five small coding tasks: 5/5 passes, about 9× slower and 48× more expensive than dsh on DeepSeek V4-Flash (warm second pass). The pixel-art widget brief took three attempts.
My AI Agent Built a Benchmarking Tool for My AI Agents
I asked Hermes, my coding agent, to build a CLI that benchmarks my local models. It came back with bench-llm — 1,660 lines of Python against LM Studio. A year of using it turned it into CrucibleForge: 251 cases, 158 of them hard, graded by execution.
InfoForge: Offline Wikipedia as a RAG Source for My Local AI Agents
All of English Wikipedia — 19.19 million entries, 52.69 GB, no internet — sitting behind a search index my AI agents query in milliseconds. Every answer cites its articles, and when it can't find one it says so instead of guessing.
StudioForge: A GPU-Only LLM Server That Replaced LM Studio on My Rig, Run Remotely by My Agents
I wrote the model server I wanted: an OpenAI-compatible llama.cpp gateway on port 1234 that plans VRAM before it loads, refuses a model rather than spilling it to the CPU, pins the ones that must stay warm, and publishes 29 MCP tools so an agent on another machine runs the whole rig. Free download, source on GitHub.
OpenClaw Model Manager: a GUI so I stop hand-editing model routing JSON
A GTK4 desktop app that edits the config deciding which AI model every agent in my local stack talks to. Safe writes, dead-model checks, and one-click Claude Code rerouting without hand-written systemd files.
ArkVault: A BorgBackup GUI Where the Drive Rebuilds Your Linux Box
A free GTK4 front end for BorgBackup that writes a guided restore kit onto the drive itself, so a bare Linux install can rebuild your dotfiles, SSH keys, and services — I ran the restore drill for real.
MailForge: A Local-LLM Email Assistant That Reads Your Mail, Writes the Reply, and Never Sends It
A self-hosted email triage assistant: an IMAP listener, a local model, and a drafts queue. It classifies, labels and writes the reply — then stops. Sending is a human click, always. Free, MIT, download below.
Browse by type
AI & Local LLM
Running AI on your own hardware — local LLMs, agents, and image generation. No subscriptions, no cloud, no computer-science degree.
Laser Projects
Cutting, engraving, and the occasional small fire. Each post has the details, process, and settings you need to duplicate it yourself.
Tools & Downloads
Small apps and utilities I built and actually use. Free downloads for personal use, source included.