Become a sponsor to Daniel Boyd

Hi, I'm Dan.

I build useful tools, improve existing ones, and occasionally decide that creating an entire simulated world sounds like a reasonable side project. I'm a former USAF civil engineer and rehabilitation counselor. Different fields, same lesson: something can look excellent on paper and still fail the person who actually needs it. That shapes how I approach software: make it useful, make it understandable, and check that it actually works.

What's on the workbench

• Audio tools: My work on CUETools 2026, CUETools for Linux, and LAME v4 builds on established open-source projects, with work spanning audio verification, repair, encoding, modernization, and the occasional attempt to turn perfectly good computers into space heaters.

• Human compatibility: I'm involved in Northgiven, a cross-platform web, iOS, and Android project aimed at helping people make better connections before chemistry, attachment, and wishful thinking start doing all the talking. It draws on relationship research, psychometrics, and what we know about human compatibility to look for stronger groundwork between people from the beginning. The goal isn't to mathematically manufacture love. It's to improve the odds that two people aren't discovering six months later that the foundation was incompatible from day one.

• More accountable AI coding: Anti-Dark-Code Skill is a tool-discipline system for AI coding agents. It grew partly out of working across projects like these, plus several others that aren't public yet. When AI can produce thousands of lines of code frighteningly fast, "it seems to work" becomes a pretty terrible engineering standard. Anti-Dark-Code is about constraining scope, keeping important checks deterministic, exposing assumptions, and making AI-generated work easier to inspect and verify.

• Living-world simulations: I'm building Chronicle Engine, a persistent simulation and game-engine project with autonomous characters, procedural skills and systems, crafting, weather, disasters, magic, and eventually multiplayer. One of the stranger goals is letting a player inhabit an NPC for a while, log out, and hand that character back to the simulation without the world politely freezing until the human returns.

• And a few things I can't talk about yet: There are other software and research-heavy projects in various stages of development. Some may become public. Some may not. All of them contribute to the same recurring obsession: building complicated things without letting complexity quietly become chaos.

The unglamorous work matters

A useful project is more than its most impressive screenshot. It is also the bug someone finally reproduces. The documentation that saves the next person an afternoon. The build that works on another machine. The improvement to an older tool that still does something worth doing.

That's work I care about.

I use AI heavily. It does not get diplomatic immunity. Generated code still needs testing, understandable behavior, and someone willing to question a very confident answer.

“It worked once on my machine” is a milestone, not a quality standard.

Why open source?

Because a useful tool can become even more useful when people can inspect it, adapt it, and build on it.

My work benefits from developers who shared theirs. I want my contributions to give the next person a better starting point, too.

Plan broadly. Build narrowly. Test reality.

Thanks for taking an interest in what I'm building.