Much of the hardest work of creating effective data products in the enterprise is not in the complexity of the algorithms applied but in effective design and integration into downstream systems. Hilary Mason (Cloudera) shares a process for repeatedly creating effective AI products, from idea through process to specific design considerations, and explains how architecture and algorithmic choices can support or hinder this process.

This session was recorded at the 2019 O'Reilly Artificial Intelligence Conference in New York.