June 2026
Spotify Pulse Analytics
AI-Powered Music Intelligence Dashboard
Overview
An ML-powered analytics platform that predicts song popularity and surfaces music trends across 10,000+ tracks via interactive executive-grade KPI dashboards.
Problem Statement
Music labels and artists struggle to identify hit potential early. Raw streaming data is voluminous and hard to interpret without specialized tooling.
Key Highlights
ML pipeline with Scikit-Learn predicting song popularity at 87% accuracy across 10,000+ track records
Interactive KPI dashboards with real-time filtering by artist, genre, and tempo — reducing data exploration time by 65%
Predictive analytics outputs delivered as executive-grade BI visualizations with CSV export for downstream business reporting
Architecture
Streamlit single-page app with a Scikit-Learn prediction layer, SQL-backed data store, and Plotly chart components for real-time filtering and visualization.
Challenges
Balancing model accuracy against interpretability for non-technical stakeholders, and designing a dashboard layout that communicates both raw metrics and predicted outcomes clearly.
Learnings
End-to-end ML pipeline integration with a live dashboard, translating model outputs into stakeholder-friendly business narratives.