NicotinD
A self-hosted music experience designed around one continuous path from finding a track to listening to it.
Problem
Self-hosted music listeners move between disconnected tools to search, acquire, organize, and play music.
Context
The underlying services were capable, but their separate interfaces made a simple listening goal feel like infrastructure work.
Discovery
Mapping the end-to-end journey showed that tool boundaries—not missing features—created most of the friction.
Hypothesis
A single experience could hide service boundaries while still giving advanced users control.
Options
I considered a thin launcher, a shared dashboard, and a fully unified workflow.
Decision
I chose unified search, optional acquisition, and cross-device playback behind one product language.
MVP
The first version deliberately left out social discovery and advanced library management.
Execution
I owned the concept, experience, architecture, and implementation, using working software to test each decision.
Outcome
The result turns a multi-service process into a coherent workflow where newly acquired music becomes immediately playable.
Learning
Integration is a product problem: technical orchestration only matters when it removes effort for the user.
