Miniko

↤ Work / Miniko

Miniko is a desktop application that lets creators build, customize, and rig expressive 2D streaming avatars without commissioning an artist or wrestling with rigging software. I lead the product design end to end, from the core creation flow to the visual system and motion.

Role: Owner and Product Designer
Timeline: 2024 — Now
Tools: Figma, Illustrator, Tauri, After Effects

The problem

Creating a streaming avatar is normally expensive, slow, and technical. Most creators either pay hundreds to thousands for a commission or skip avatars entirely. Miniko removes the middleman, allowing you to pick parts, customize, and go live within seconds.

Old way vs. the gap — commissioning an artist or rigging it yourself both dead-end where most streamers give up; Miniko bridges the gap in seconds with no artist or rigging.

Exploration

I started with the flow, not the screens. I mapped the path a creator takes to go live, then referenced existing avatar programs to define Miniko's structure and the features it needed at a minimum.

Miniko Ver. 1 — three early screens: home with mic-reactive states, the character editor, and the in-app store. — The first iteration, built straight from that flow. What started here became the core of Miniko today: mic-reactive states, live customization, and asset selection.

During exploration I mocked up everything: control schemes, color systems, an in-app store, icon sets. So many ideas pulled me in different directions. I kept telling myself I was protecting the simplicity, but with this much creative freedom, it was easy to get carried away.

A board of explored Miniko directions — shop mockup, color palette mockups, sold designs and icons, and a hands control panel — most later cut.

I shipped a version — then pulled it

I launched a version of Miniko on Steam, and then pulled it. The app had gotten too complex because I'd built it store-first. That one decision created too much navigation for users to wade through (coin shop, purchase prompts, dozens of states), until making a simple avatar no longer felt simple.

The Miniko Ver. 2 board — around twenty screens including the editor, in-app store, coin shop, purchase prompts, and loading and error states — the sprawl that shipped on Steam before being pulled. — Designs were feature-bloated, store-first, off-brand.

Simplicity was the valuable lesson taught here. I'd been too focused on how the design worked around the software's pricing and the features I wanted to implement. So I went back to the drawing board and started from ground zero, cutting back every unnecessary feature with the core idea in mind: a simple focused editor.

A single, focused creation flow

At this point, Miniko's customization was scattered: a complex editor, an overwhelming in-app shop, and fragmented controls. Feedback from my Discord community and Steam made one thing clear, creators wanted every asset available up front. So I collapsed it all into one cohesive editor with fewer interactions, less clutter, and smoother transitions between outfits, colors, and states.

The home screen everything runs from: live avatar, expressions, effects, mic, and outfit, all in one view.

The Miniko editor redlined — padding, gaps, and control sizes annotated across the screen (8, 16, 40, 90px), showing one consistent spacing system.

The wardrobe, where creators dress their avatar: a color picker, an item selector, a preset palette of quick colors, and item categories down the side.

Down at the control level, some refinements I made:

The trickiest was the face-tracking calibration icon. It had to communicate a complex idea, aligning your camera to your face, in a single glyph small enough to sit inline with the rest of the controls.

The face-tracking calibration icon refined across several iterations — from a busy early glyph to a single clear mark that reads at small size. — I iterated through several versions, simplifying the shape each pass until it read instantly at icon size without needing a label.

The Miniko outfit dropdown before and after — a plain dropdown, then with edit, duplicate, and delete, then with an added hotkey bind. — The outfit control, refined — a plain dropdown gained inline edit, duplicate, delete, and a hotkey bind.

The Miniko color picker before and after — a bare picker, then with an eyedropper added, then with a saved-swatch palette. — The color picker, refined — from a bare picker to an eyedropper, then a saved-swatch palette for reusing colors fast.

A consistent component system

The Miniko component set — dropdowns, sliders, the color picker, parts grids, icons, and their states, laid out together. — Components were built to be one modular kit: dropdowns, sliders, the color picker, parts grids, icons, each with its states. Modularity allowed me to rearrange them into different mockups and quickly see which layout worked best.

Miniko's system colors and type scale — named color tokens with hex values and an Inter type ramp. — The tokens behind Miniko's brand: system colors and an Inter type scale that keep every screen feeling like one product.

Motion that communicates

Simplifying the interface also meant leaning on motion to guide the eye. I wanted real-time feedback, with assets and colors updating instantly, while keeping the experience clear without on-screen instructions.

Beyond the in-app UI motion, I moved into After Effects to build an animated logo and UI prompt, extending that motion language into the brand itself so Miniko feels modern and effortless to use.

Bringing Miniko to the web

Miniko's software is only half the story as I had also designed the site that sells it. The landing page has one job: turn a curious streamer into a download. I led with the avatars themselves, kept the path to download short, and carried the product's visual language straight through so the brand reads as one thing.

Desktop and mobile — avatars lead, the Buy button sits one click from the top, and the product's look carries straight through.

Impact

After finishing Miniko, I was finally able to launch it. That's when it hit me: Miniko isn't a prototype, it's a product strangers pay for. As of writing this case study, it has earned $679 across 29 paid orders, and it's still growing.

Miniko's order dashboard — 29 paid orders and $679.10 in cumulative revenue, with a table of paid transactions, customer names blurred. — Real, paid orders — $679 across 29 sales. Customer names blurred for privacy.

Reflection

Building Miniko made me an all-in-one researcher, designer, and front-end engineer. Unlike a corporate setting where there is friction and many pass-throughs for a small change, I was in complete control of the project. That freedom taught me a valuable lesson in how to cut scope hard, which in turn kept the creation flow simple for creators. Moving forward, I'll carry this lesson beyond Miniko, applying it to my career and every product I work on.