Senior AI Scientist - SAP Globalization AI

Description

The role is Senior AI Scientist on the SAP Globalization AI team, within SAPGLO E-Invoicing Localization.

The team spans 40 locations and builds AI agents, intelligent tools and the infrastructure to scale them, applying AI to coding, research, automation and software design to speed up compliant solutions for markets worldwide.

This is an active build phase, and learning, coaching and knowledge sharing are part of how the team works.

The company is a leader in enterprise applications and business AI, serving finance, procurement, HR, supply chain and customer experience for over 50 years.

Responsibilities

• Set the scientific methods and evaluation standards for generative AI, RAG and agentic systems, from hypothesis and experiment through to production.
• Own evaluation strategy and quality standards, defining hypotheses, baselines, success metrics and acceptance criteria.
• Build benchmarks and evaluation methods for retrieval quality, grounding, factual consistency, safety and task completion, using responsible, privacy-aware datasets.
• Run systematic error analysis to find root causes in data, retrieval, ranking, context, prompts, models, tools or agent orchestration.
• Research embeddings and hybrid search, with a focus on reranking, model selection and agent architectures, using controlled experiments and both offline and online evaluation, including validated LLM-as-a-judge methods.
• Design, build and run production AI capabilities, evaluation services and data pipelines, monitoring quality, drift, latency and cost.
• Work with product managers, domain experts and engineers to put validated approaches into production.
• Share scientific and engineering practices across SAP Globalization and mentor technical colleagues.

Requirements

• Work as an applied AI scientist with solid software engineering skills and a record of running production AI systems.
• Communicate well, collaborate easily, and think creatively while learning fast enough to keep up as the role changes.
• Hold a degree in Computer Science, AI, Machine Learning, Data Science, Statistics or a related field, or have equivalent practical experience. A master's or PhD is an advantage.
• Write production-quality Python, and have built APIs, reusable libraries, automated pipelines and cloud-deployed components, including CI/CD and automated testing.
• Have a grounding in applied statistics, machine learning, deep learning, NLP and information retrieval, plus hands-on work with LLM applications, RAG, embeddings, semantic retrieval and agentic workflows.
• Have designed evaluation datasets, benchmarks and quality metrics, understand offline and online evaluation and statistical significance, and can trace AI failures to their root cause.
• Use AI-assisted tools and automation to build better solutions, and make sure AI-generated outputs are high quality, safe and aligned with company goals.
• Fluent English is required.

Conditions

Equal employment opportunity applies, and accommodations are available for applicants with disabilities through the Recruiting Operations Team at.

Successful candidates may need to pass a background verification with an external vendor.

Use of AI in the hiring process must follow the company's ethical AI guidelines, and breaking them may lead to disqualification.

Benefits

• Flexible working models and a focus on health and well-being.