SAP TechEd showcases generative AI capabilities for developers and business users, including AI-powered code generation and vector database processing.
SAP SE will flex its artificial intelligence muscle at its TechEd event this week by announcing a long list of generative AI capabilities aimed at developers and business users.
The headline act is SAP Build Code, a version of the enterprise resource planning giant’s Build low-code development platform for software professionals that folds in AI-powered productivity aids optimized for Java and JavaScript development. It uses the generative AI copilot Joule that the company launched in September to generate code for data modeling, application logic and test script creation.
SAP said SAP Build Code is aimed at improving cooperation between businesspeople who use Build and professional developers. It connects to SAP and third-party software data and systems and uses simplified application program interface management to let developers create apps and processes without losing business context. The new development tools also interoperate with the ABAP Cloud development model for cloud-ready business applications and features governance and lifecycle management capabilities.
The second major announcement this week is support for vector database processing in HANA Cloud. HANA is a column-oriented in-memory database management system that runs advanced analytics alongside transactions. A vector database is a specialized type of database designed to store, manage, and query vector data, which is an array of numerical values that are often used to represent things like features in machine learning, spatial coordinates in geographic information systems or multidimensional data.
Search similar
Vector databases enable searches to be performed based on the similarity between vectors rather than exact matches. SAP gave the example of searching for suppliers based on the language in their contracts. The capability is widely used in AI applications like image recognition, natural language processing and recommendation systems.
“This is important because large language models all have a cutoff date,” said SAP Chief Technology Officer Juergen Mueller. “They were trained with certain data but they don’t have the latest data or company-specific data. Vector databases can give more context. We are getting closer to the holy Grail of generative AI where you can combine the strength of large language models with real-time company data.”
Another new initiative aimed at AI development is AI Foundation, a toolkit developers can use to create AI- and generative AI-powered extensions and applications atop the SAP Business Technology Platform. BTP provides enterprise management capabilities such as database security, encryption and governance. SAP said AI Foundation includes everything developers need to create applications such as intelligent document recognition and machine translation on BTP, including prepackaged AI services, access to the most popular large language models, vector database capabilities, and AI runtime and lifecycle management.
Also being rolled out today is a new role-based certification for back-end developers using the ABAP Cloud development model. Associated learning resources are intended to help developers build the skills needed to create cloud-ready extensions that align with SAP’s clean core practice of minimizing customization and modifications to the core ERP system. New learning resources covering ABAP development tools on SAP BTP and SAP S/4HANA are now available on the SAP Learning site.
Dozens of other announcements are planned for this week, including enhancements to the SAP Discovery Center, open-sourcing of the SAP Cloud software development kit for Java, a new cloud logging service, a containerized runtime option for SAP Integration Suite, and generative AI capabilities for developers who want to build AI extensions with the ABAP programming language. A full list of TechEd announcements is available on the TechEd News Guide.
