Work

About

Contact

Work

About

Contact

Work

About

Contact

Case study

Acrylic

An AI music licensing platform shaped from an open brief into a build-ready MVP direction.

Founding product designer · 2026

Role

Founding product designer

Timeline

2026

Scope

Product strategy, IA, UX/UI, design system, handoff

Outcome

Build-ready MVP direction

Role

Founding product designer

Scope

Product strategy, IA, UX/UI, design system, handoff

Timeline

2026

Outcome

Build-ready MVP direction

The challenge

Acrylic set out to make licensing music feel as fast as finding it. The hard part sat underneath. Licensing carries real constraints: usage rights, business rules, and multi-step approvals. And the teams choosing music move fast, and want to trust what they pick before they commit. When I joined, that product space was still abstract. The work was to give it a structure that could carry that weight without slowing the people using it.

Acrylic set out to make licensing music feel as fast as finding it. The hard part sat underneath. Licensing carries real constraints: usage rights, business rules, and multi-step approvals. And the teams choosing music move fast, and want to trust what they pick before they commit. When I joined, that product space was still abstract. The work was to give it a structure that could carry that weight without slowing the people using it.

What I inherited

An early-stage product with conviction but few settled decisions. There were strong product ideas and a technical team building toward them, but no agreed information architecture, no buyer-side flows, and no shared sense of how licensing should actually read to someone in the middle of a choice. The catalog could have been enormous, the rules were genuinely complicated, and the team needed something concrete enough to build against. I came in to turn that open space into defined structure.

An early-stage product with conviction but few settled decisions. There were strong product ideas and a technical team building toward them, but no agreed information architecture, no buyer-side flows, and no shared sense of how licensing should actually read to someone in the middle of a choice. The catalog could have been enormous, the rules were genuinely complicated, and the team needed something concrete enough to build against. I came in to turn that open space into defined structure.

My role

I owned the buyer-side experience end to end: how someone searches, evaluates a track, organizes options with a team, and moves to licensing. I set the information architecture, designed the core flows and interface, and built the design system the engineers would work from. I worked directly with the founder and technical team, translating product intent into flows and patterns concrete enough to build. Where the product was undecided, I made the call and showed the reasoning behind it.

I owned the buyer-side experience end to end: how someone searches, evaluates a track, organizes options with a team, and moves to licensing. I set the information architecture, designed the core flows and interface, and built the design system the engineers would work from. I worked directly with the founder and technical team, translating product intent into flows and patterns concrete enough to build. Where the product was undecided, I made the call and showed the reasoning behind it.

Key decisions

Decision

Lead with search, not the catalog.

Reasoning

People come to Acrylic to find the right track and move on, not to browse a library. Putting search first kept intent at the center and stopped sheer volume from becoming the product’s center of gravity.

People come to Acrylic to find the right track and move on, not to browse a library. Putting search first kept intent at the center and stopped sheer volume from becoming the product’s center of gravity.

Tradeoff

It gave up the showcase value of a large browsable catalog, a bet that confident retrieval matters more than exploration.

Decision

Present licensing as guided choices at the point of decision, not as terms to read.

Reasoning

Teams don’t want to become licensing experts. They want to trust the pick. Surfacing only the few signals that actually change a decision built that trust without turning the flow into a contract.

Tradeoff

Deciding what to surface meant deliberately holding back edge-case detail, accepting some loss of completeness in exchange for everyday confidence.

Decision

Build the experience around projects, so choosing a track could be a team activity.

Reasoning

A track is rarely chosen alone or in one sitting. Organizing candidates into projects matched the real rhythm of a fast creative pick followed by slower internal review.

Tradeoff

It added structure to an MVP that could have shipped a simpler single-user flow sooner, a bet that collaboration was core rather than a later addition.

Selected artifacts

A few of the artifacts behind the decisions above.

Research board: the discovery and evaluation patterns I scanned before designing anything. What mattered here was how teams actually decide on a track, not how catalogs look.

Competitor scan: where existing tools served buyers well and where they left them guessing. The gaps, more than the features, set the direction.

Site map and core flow: the path from search to licensing. Notice the project layer threaded through it, where collaboration lives, rather than bolted on at the end.

Outcome

The work established a coherent buyer-side direction for the MVP and a design system the team could build against. It gave an abstract product a concrete shape: a search-led entry, evaluation and licensing flows that read as decisions rather than paperwork, and a project structure built to scale as the team grew. By the end, product intent, interface, and engineering were working from the same foundation, and the MVP was ready to move into build.

The work established a coherent buyer-side direction for the MVP and a design system the team could build against. It gave an abstract product a concrete shape: a search-led entry, evaluation and licensing flows that read as decisions rather than paperwork, and a project structure built to scale as the team grew. By the end, product intent, interface, and engineering were working from the same foundation, and the MVP was ready to move into build.

The work was never decoration. It was judgment about what to simplify, what to signal, and what to leave out, and the willingness to stand behind each call.

← Back to work

← Back to work

© 2026 Arman Musaji

Email:

armanmusaji@gmail.com