Blog

Building for Autonomy

Thoughts on agent-first infrastructure, self-hosted systems, and owning the stack.

Filtering by: Engineering × Clear
#machin #ai-agents #open-source #agent-first #ai-engineering #devtools #peage #self-hosted #agents #cli #compiler #security #automaintainer #dogfooding #mfl #remotecmd #infrastructure #performance #roam #supercli #x402 #a2a #agent-economy #analytics #compilers #concurrency #devops #game-dev #grange #hart
Engineering · August 2026

I benchmarked my language against Rust and Zig, and deleted my best number

I set out to prove my language beats Rust and Zig. The first thing I found was my own benchmark measuring the order I ran things in — turning a claimed 20-25% win into a tie. Here is what machin actually wins, what it loses, and the 45ms speedup I wrote and then threw away.

#machin #benchmarks #rust #zig #compilers
Read more
Engineering · August 2026

Where a tool should stop, and ask

I built a sprite pipeline for my 2D engine and hit a wall: to swing a leg, you must first know which pixels are the leg. That is not a measurement. So the tool stops and asks the agent — and grades the answer. Updated with the parent-joint fix.

#machin #agents #game-dev #pixel-art #cli #tooling
Read more
Engineering · August 2026

15 Lines of Bash Beat My 3,000-Line Search Engine

I built a Go search engine with TF-IDF, graph theory, and bidirectional stemming. It lost to a shell script embedded in a 79 KB binary. Then I made the Go engine learn from the shell script. Here's what happened.

#memgraph #supercli #skills.match #search #irony #ai-engineering
Read more
Engineering · July 2026

A 7-billion-parameter LLM at the speed of a 1B — in pure MFL

I ran OLMoE-1B-7B — a 7-billion-parameter Mixture-of-Experts model — in pure MFL, token-identical to the fp32 reference, streamed from disk, at 14.5 tok/s on a laptop. The disruption wasn't in the engine; it was in the model.

#machin #llm #moe #inference #pure-mfl
Read more