Рет қаралды 332
Presented by: Manasi Vartak, Founder & CEO at Verta
The most popular term in machine learning and data science in recent times is probably “MLOps”; it has been used to describe everything and nothing about the process of creating and deploying models. However, the fact of the matter is that building AI-enabled products remains a major challenge and requires more than just software or model development practices; it requires a unique combination of practices from software development, ML development, and data management. Loosely termed DevOps, MLOps, and DataOps respectively, only together can these three areas enable rapid development of intelligent products.
Drawing upon experiences bringing dozens of AI-enabled products to market, this talk will discuss what it takes to ship AI-enabled products; demystify the terms DataOps, MLOps, DevOps; and provide best practices for structuring teams to support intelligent software development.