Balancing Data Sharing Concerns: Code as the Bridge

Maria highlights their organization's approach, stating that they do not share data or models directly. Instead, they emphasize sharing the code by maintaining centralized repositories for each project type. Basak adds that when developing models, they consider the possibility of multiple brands utilizing them. They ensure configurability and parameterization so that the essential package acts as a Python package imported into each brand's repository, enabling model training and deployment. This approach allows for collaboration while addressing data privacy and intellectual …

brands bridge code data data sharing highlights multiple project type