People developing enterprise applications have a different mindset. Unlike data science projects, software engineers typically know upfront what they need to build the system. They also know whether they have to live with certain assumptions and policies. For example, the technology stack may already be known when the project starts. Other factors that may already be familiar include the system architecture that will be used, which cloud vendor will be utilized, what database the application must integrate with, and so on.

Enterprise applications typically require a rich business domain object model. Data objects are created, manipulated, and transferred to different layers of the application. The system architecture may include a user interface, a middle tier, and a database backend.

Enterprise applications also tend to require a high level of integration with other systems. For example, a trading system used by an investment firm is typically hooked up to an accounting system, a trade-settlement system, a reporting system, and so on. As such, these applications are often designed to handle both data at rest (for example, data stored in a database) or data in motion (for example, data being streamed to another system). Furthermore, data movement may happen in real-time or as an overnight batch process.