This role focuses on building AI-native, HANA-optimized SAP applications and extensions. You will combine deep ABAP on HANA engineering with agentic AI patterns, including LLMs, tools, retrieval, and evaluation. The emphasis is on delivering high-performance, cloud-ready, production-grade solutions using HANA pushdown principles and modern ABAP. You'll leverage governed AI capabilities that can reason, retrieve context, and assist or act within enterprise SAP workflows, without the need to train foundation models from scratch.

Design and develop ABAP solutions optimized for SAP HANA using pushdown principles to maximize in-database processing. Build efficient data models and access patterns with CDS views, associations, and annotations. Implement performance-critical logic using AMDP/SQLScript where appropriate. Develop modular, testable, and maintainable ABAP code aligned with clean architecture and enterprise SDLC. Build services and APIs using modern paradigms like OData, ensuring stability and versioning. Model enterprise business entities using CDS semantics for transactional and analytical consumption. Design reusable data artifacts to support downstream consumers and AI retrieval layers. Build AI-enabled SAP experiences integrating ABAP services with LLM capabilities, focusing on tool calling, retrieval-grounded outputs, and deterministic fallbacks. Implement prompts and context assembly strategies to reduce hallucinations and enforce enterprise safety boundaries. Build automated tests for ABAP services and data artifacts. Establish evaluation harnesses for AI behaviors and introduce guardrails for AI outputs. Instrument services with logs, metrics, and traces for production support. Design robust error handling for enterprise flows and support incident response.

Requires strong hands-on expertise in ABAP Development on HANA, including performance tuning and pushdown techniques. Proficiency with CDS views and modern data modeling practices is essential. Experience with AMDP/SQLScript for performance-critical scenarios is necessary. A solid software engineering discipline, including clean design, a testing mindset, CI/CD orientation, and production readiness, is expected. AI-native build capabilities, such as LLM integration, retrieval grounding, tool orchestration, evaluation loops, and safety boundaries, are a must-have. A minimum of 3 years of experience and 15 years of full-time education are required.