Research Papers
In-depth analysis of Data Engineering pipelines and Deep Learning models for music recommendation.
Data Engineering
Scalable Music Data Pipelines and Feature Stores
Data Engineering Research
This research focuses on the data engineering backbone required to support music recommendation systems at scale.
ETL PipelinesExternal APIs IntegrationData WarehousingStructured & Unstructured DataBatch vs Streaming
Outcome:A unified analytical data warehouse capable of supporting downstream machine learning workloads and exploratory data analysis.
Machine Learning
Audio Representation Learning for Music Recommendation
Machine Learning & Deep Learning Research
This research explores how deep learning models understand music directly from audio signals.
Spectrogram-based LearningTransformer-based Audio ModelsPretrained Embeddings (OpenL3, CLAP)Vector Similarity SearchContent-based Recommendation
Outcome:A recommendation pipeline that compares songs using learned audio embeddings rather than relying solely on metadata.