About ENOSYS

About ENOSYS

ENOSYS, also known as “The Sweeping Monk” (扫地僧).

Born in Haidian, Beijing, an alumnus of Haidian Huangzhuang No. 1 High School, and currently based in Silicon Valley, California.

I’m a software engineer by trade. For over two decades, my work has centered on system software, infrastructure, and security. Spend enough years writing code, and you inevitably pick up a few stubborn engineering habits: interfaces should be exactly what they ought to be, never add an extra layer of abstraction when one less will do, and if there is unnecessary complexity, you instinctively want to tear it down. Over time, peers have occasionally joked about me being an “Internet old fart.”

Beyond terminals and systems, though, I’ve always had an interest in sound and music. Compared to code, music is noticeably less reasonable—which is probably why it’s so much fun.

People often ask where the name “ENOSYS” came from. There is no deep philosophy behind it, let alone any brand strategy; it was just a name I threw together many years ago. At the time, I didn’t even realize that for people who don’t write code, these letters are quite a tongue-twister (roughly pronounced eh-NOH-sis or ih-NOSS-iss, IPA: /ɪˈnɒsɪs/).

In Unix/POSIX system programming, ENOSYS is an uncommon error code indicating that the requested function is not implemented or not supported. When a program hits it, it is essentially the operating system shrugging:

“Don’t even bother; this feature doesn’t exist.”

Using it as the name for a music project turned out to be unexpectedly fitting: I’ve never had formal training in composition, my music theory is rudimentary, and my instrument skills are far from professional. Through an engineer’s lens, this musical subsystem lacked plenty of features right from the start—ENOSYS everywhere.

Fortunately, generative models and audio tools have advanced rapidly in recent years. If it’s missing in the kernel, you just patch a layer in user space. What I cannot play, models help play; what I cannot arrange, I tweak by ear. In the end, I’ve managed to piece together things that I actually enjoy listening to on repeat.

Music & Experiments

My initial reason for exploring AI-assisted music was simple: it just seemed fun.

I’ve always had a soft spot for tools that get computers to do things they weren’t originally supposed to do. That was true when I was hacking on code in my younger years, and later with images and audio—it’s largely the same urge. To me, algorithms and compute power are not only tools for solving engineering problems, but also an intriguing creative medium.

Currently, I independently produce and release music under the name ENOSYS the Sweeping Monk. These releases include melodies, arrangements, and performances assisted by AI, alongside experiments heavily driven by generative models from start to finish. There is no strict formula for the genres: ranging from straightforward instrumentals and game music to electronic tracks, all the way to weird, playful experiments that are basically just having fun.

Present Status

Currently based in Silicon Valley, married with one daughter.

By day, I still earn a living by writing code, maintaining systems, and resolving all kinds of strange issues. In whatever time is left, I make music, tweak audio, or tinker with things that nobody asked for in the first place.

This project has no clear finish line, nor any grand ambition.

Outside of work, everyone needs a corner to do things that aren’t necessarily useful, but genuinely enjoyable.