Research on data management for machine learning at the University of Amsterdam (as part of INDELab). Checkout our recent and ongoing projects:

• mlinspect allows the instrumentation and inspection of native ML pipelines written in Python with pandas, sklearn and keras.

• serenade is a low-latency session-based recommender system deployed in production at bol.com, the largest e-commerce platform in the Netherlands.

• caboose contains implementations of state-of-the-art kNN models for next-basket recommendation which can "unlearn" user data in milliseconds.

• jenga is an experimentation library that allows data science practititioners and researchers to study the effect of common data corruptions (e.g., missing values, broken character encodings) on the prediction quality of their ML models.