Supervised (self-titled)

This was 2016, when machine-generated images weren't as common as in today's GPT era, and when a computer accidentally generated interesting images without a prompt. For Supervised's first and self-titled album, I was visually a bit lost, having spent a lot (maybe all of it!?) of my creative energy recording, mixing, and collaborating with other musicians. But my designer self wanted more — to find a visual language that felt right, that harmonized with the name, with the music.

I really enjoy spending way more time than I can justify creating concepts—and going through the frustration that is in refining, refining, refining—until I find the right shape. The one that shines more than a dozen other alternatives.

At the same time, I love learning new tools—even with the boring parts: figuring out software interfaces, waiting, crashes (hey, Blender on Mac!), and the messy, unfinished work that comes from that initial learning obsession. What I love about that phase is that I'm giving ZERO IMPORTANCE to the output, and focusing on understanding what the tool may offer, which means that I get A LOT of outputs. And 99.9% of these will always be unborn creations in my backup folders (sorry if that’s an unsettling analogy!).

For this particular cover, we are looking at the remaining 0.1% that grows and becomes directional to so many future ideas. This somewhat simple randomness happened during the 1,823,918,239,812th time that Photoshop (yes, I'm old) just stopped working, without any notice. Interestingly, as I clicked through a frozen canvas, it was refreshing (and crazy) to see seemingly random, glitchy compositions. Was it the muse in the machine? I will never know! But at least I thought about screenshotting some of it and putting it into the archives.

However, it was just a screenshot I had taken from a glitchy moment, and it demanded some work (as it always does), re-texturing amd re-coloring, while still preserving the machine-made composition.

I love the tension that emerges, knowing I wasn’t fully responsible for designing this cover—but somehow I was! It reminds me that even my most obsessive “refine, refine, refine” moments can yield unexpected collaborators—software included.

Why Supervised?

When chasing the name for my band (or musical project), I was wondering about the constant supervision that we’re all bound to, and contrasting that with the days when I started making music as a teenager, in a small city, where kids and teens (everybody?) weren’t constantly watched. Supervised was one of the ideas that came to mind. I also realized that some people would think about Supervised Learning, a machine learning technique to train “AI Models”.

Looking back, the machine-generated visuals and the meaning of the word Supervised now feel like a perfect match.