IBM Watson Studio is a data science and machine learning platform for building, training, and deploying models within IBM's Cloud Pak for Data environment. IBM introduced it in March 2018 at its Think conference as a rebrand and expansion of the earlier Data Science Experience (DSX), which had launched in 2016, folding DSX's notebook tooling into a broader AI development suite.
The platform centers on collaborative Jupyter and Zeppelin notebooks, a visual model-builder called AutoAI that tests multiple algorithms against a dataset automatically, and Watson Machine Learning for deploying trained models as APIs. Teams work in shared projects where data engineers, analysts, and reviewers open the same assets, and compute is metered in Capacity Unit Hours (CUH) rather than flat per-seat licenses, so a data science team running heavy training jobs pays differently than one doing exploratory analysis.
Watson Studio targets enterprise data science teams already inside IBM's cloud ecosystem rather than solo practitioners comparison-shopping against Databricks or SageMaker. A Lite tier gives one user 10 CUH a month at no cost for evaluation, while the Professional tier removes the user cap and bills for compute consumption at rates IBM negotiates per account rather than publishing a flat sticker price.
