MusicMind Agent Platform

An autonomous multi-agent platform that enriches music data by fetching from multiple sources in parallel and merging results into an Aerospike graph database.

What It Does

Given a song and artist, the platform fans out requests to multiple data sources simultaneously -- Spotify, Last.fm, MusicBrainz, and web scraping -- then merges, deduplicates, and persists the enriched data as a knowledge graph of music entities and relationships.

Key Features

• Multi-agent orchestration -- parallel data fetching with conflict resolution and completeness scoring
• Graph data model -- songs, artists, albums, labels, instruments, venues, and concerts stored as nodes and edges in Aerospike
• Self-improvement engine -- tracks data quality per source, processes user feedback, and schedules re-enrichment of stale data
• React frontend -- graph visualization and social features
• API -- FastAPI with JWT auth, rate limiting, CSRF protection, and input validation
• Caching -- Redis with configurable TTL and LRU eviction
• Observability -- distributed tracing via Overmind

Tech Stack

Built With

• python
• react