SoundVent needed a social layer purpose-built for music culture — one that could model relationships between people and entities, power feeds and profile walls, and generate structured engagement data, and instead of bolted-on comments and likes, it required a unified, entity-aware system at the core of the platform.

I built a shared social graph that powers follows, posts, comments, likes, and shares across feeds, profile walls, discovery, and overlays, and all social activity is funneled into a unified model that supports both rich user experiences and deeply queryable analytics for creators.

A Supabase-backed, entity-aware social graph sits at the core of SoundVent, with a dedicated feed engine, engagement pipeline, and clips layer all sharing the same model, and strict row-level security and typed, composable services keep the system reusable, secure, and consistent across every surface.

The social experience feels familiar yet music-native, with feeds, profile walls, inline posting, and threaded conversations woven directly into existing flows, and dark, minimal visuals and thoughtful overlays keep interaction fluid and expressive without pulling users into a separate “social mode.”

Key challenges included modeling people and entities in one graph, balancing global discovery with relevance, handling rich media, keeping engagement state accurate, and preserving performance in heavy feeds, and these were solved with entity-aware identity modeling, a flexible feed engine, normalized media, idempotent engagement endpoints, and performance-focused UI patterns.

The social system turns SoundVent into a living network where artists and fans connect, interact, and build relationships, rather than just consume content, and it also establishes a reusable, entity-aware pattern for future features, proving out the platform’s core architecture and UX philosophy.