ML model adoption has increased over the past few years, but in many organizations there is still a wide gap between model development and model use in a production environment. This is due in large part to the difficulties in establishing pipelines and integrating models with existing workflows and legacy applications. This gap becomes even wider as more models are added.
In this talk, we will present an approach which automatically builds pipelines and establishes a scalable and intuitive solution for constructing a highway between data, machine learning models, and end user applications, ensuring that everybody in the organization can be seamlessly empowered with actionable information – presented how, when, and where it will be most impactful. This concept extends the Reverse ETL paradigm by enabling the connection of multiple data sources and machine learning models with the user interface of currently existing SaaS or internally-developed applications, simplifying the data engineering workflow required to support the end consumers of the data or model’s outputs.