Creative Technology · System Design · Brand Strategy · AI Assisted Development
Selkie
An independent creative project translating music into a visual release world.
I conceived and built Selkie as a working prototype exploring how AI could help independent musicians turn a song and creative direction into a coherent visual release system.
The music is ready. The visual campaign usually isn't.
My research pointed to more than 100,000 songs being released every day, and to many artists unsure how to brand themselves. An emerging artist I spoke with pointed to content as one of the main things people would pay for. That conversation helped focus Selkie on turning an artist's sound and direction into a connected release system.
02 / How the system works
A structured creative process. Not a black box.
Selkie organizes the track, release context, references, exclusions, and artist intent into a creative brief. Prompt rules translate that brief into three deliberately different worlds, each with its own concept, color roles, typography, image language, texture, and motion behavior.
The system proposes the creative territory. The artist chooses, combines, and redirects where it goes.
Three world comparison · Responsive single world view03 / Input logic and artist control
From context to a steerable system.
01
Give Selkie context
The artist uploads a track and adds release details, references, intended feeling, and exclusions. These inputs become constraints for the system rather than decoration around a generic prompt.
02
Compare three worlds
Prompt and workflow rules separate the response into three art direction propositions rather than treating the first generated look as the answer.
03
Mix and redirect
Through the World Mixer, the artist can keep a world intact or combine its imagery, palette, typography, and energy with another direction.
04
Shape the output
Artist visibility, image treatment, composition, palette, typography, and restrained, charged, or fractured motion modes shape the final system.
04 / Testing and iteration
The useful work happened between versions.
I tried the experience with different songs and creative inputs, checking whether the worlds felt distinct, the controls changed the direction in meaningful ways, and the outputs could work beyond the screen.
Early failure
Different prompts produced similar looking worlds.
Palettes, fonts, and image language began repeating across songs, which made the system feel stylistic rather than responsive.
Refinement
I tightened the rules separating each world and defined color, type, imagery, texture, and motion as distinct system variables.
Early failure
The artist had too little control after generation.
A strong first image was not enough. References could repeat, questions felt abstract, and early motion studies were too uniform or slow.
Refinement
I added clearer inputs, exclusions, the World Mixer, cover controls, and motion modes so the artist could redirect the result instead of restarting.
Early failure
A moodboard did not equal a release campaign.
The concept needed to move from inspiration toward assets an artist could actually use or hand to a collaborator.
Refinement
I expanded the workflow into a cover builder, a color and typography system, motion output, and a downloadable creative brief.
05 / Project walkthrough
The product, in motion.
Walk through the complete v0.13 flow from track and artist input through three world generation, World Mixer controls, responsive cover styling, motion modes, and downloadable outputs.
Each session develops into a connected release system rather than a single generated image. In the Glimmer example, the artist can refine and download the cover, export an eight second motion loop, use a color and typography system, and hand a generated brief to a collaborator.
Cover builder · Artist visibility · Image treatment · Composition
Release hues · Type system · Downloadable collaborator brief8 second motion loop preview
Cover export
8-second loop preview
Release hues
Typography system
Collaborator brief
Motion modes
07 / Identity and naming
Transformation as identity
In folklore, a selkie moves between two forms. The name became a metaphor for the product itself, an experience designed to translate sound into image while preserving the feeling that connects them.
08 / What I built
From concept to working creative experience.
I developed the product concept and built a working web prototype, including its input flow, prompt logic, three-world framework, artist controls, visual identity, motion, and downloadable outputs. I used an AI-assisted HTML, CSS, and JavaScript workflow and deployed the experience on Netlify.
Product concept and guided workflow
Prompt logic and visual directions
Artist controls, identity, and responsive interface
Motion and connected output system
Prototype development and deployment
09 / Tool stack
Tools followed the system.
The stack combined creative direction and interface design with AI assisted development. Each tool supported a specific part of the workflow rather than standing in for the idea.
GenerationOpenAI image workflows
Source materialArtist uploads + Pexels textures
DevelopmentAI assisted HTML, CSS + JavaScript
MotionBrowser based animation + export tests
DeploymentNetlify
10 / Reflection
What Selkie taught me
Building Selkie gave me a way to connect brand strategy, visual systems, interaction design, and AI experimentation in a working product. The clearest lesson was that useful creative technology needs distinct directions, meaningful artist control, and outputs that hold together as a system.