After a data science model has been developed and validated, getting business value depends on being able to reliably and securely run the model. Especially in an industry with highly sensitive data, such as the banking industry, it is crucial that the model runs in a secure environment and that the data that runs through it is safe and shielded at all times.
So, how should you go about putting a machine learning model in such a production environment? Iman Hitli and Mara Leest, who work as Analytics consultants at Avanade, will show a real-life example from a project they do at a large Dutch bank. They use YAML pipelines via Azure DevOps to deploy the model to the environments in Azure, the Python wheel library to package the model code, and a carefully thought-out testing strategy.
Join them if you want to learn how to get real business value from your data science models!