New opportunity

Senior Data Scientist- Finance & Spend, Data Labs

SAP · Bangalore, India

About this role

As a Senior Data Scientist in the Data Labs unit at SAP, you will build the semantic and contextual foundation for SAP's AI agents and assistants. You will be responsible for designing enterprise ontologies and semantic models that ground AI in real-world business data, ensuring accuracy and reliability in production. Responsibilities: Design and maintain enterprise ontologies and semantic models to harmonize data across SAP, Salesforce, Workday, ServiceNow, and IoT systems. Build AI capabilities including RAG pipelines, embeddings, vector databases, and enterprise knowledge grounding. Develop generative AI and LLM-based solutions using structured and unstructured data. Leverage SAP data models and business process semantics to ground AI solutions. Work with cloud platforms like Databricks, AWS, Azure, and GCP. Partner with cross-functional teams to translate business challenges into AI solutions. Apply machine learning and deep learning to develop and evaluate AI models. Required Qualifications: 8+ years of experience in knowledge engineering, semantic data systems, applied AI, or data science. Master's or PhD in Computer Science, Applied Mathematics, Statistics, Engineering, or a related quantitative field. Hands-on experience with enterprise ontologies, graph query languages (SPARQL, Cypher, or GQL), and knowledge of RDF vs property graph databases. Proficiency in modern GenAI systems, RAG, and vector databases. Strong Python and SQL skills with experience in ML libraries like PyTorch, TensorFlow, or scikit-learn. Proven track record of deploying AI/ML solutions in production. Experience with big data infrastructure and cloud environments. Preferred Qualifications: Deep knowledge of SAP data models and platforms (SAP Datasphere, HANA Cloud, One Domain Model). Expertise in W3C stack (OWL, RDF, SKOS, SHACL). Domain knowledge in Finance and Spend processes. Experience with agentic AI, reasoning frameworks, and multi-agent architectures.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonSQLKnowledge EngineeringSemantic Data SystemsApplied AIData ScienceOntology DesignSPARQLCypherGQLRDFProperty Graph DatabasesGenerative AIRAGEmbeddingsVector DatabasesSemantic RetrievalPyTorchTensorFlowscikit-learnDatabricksAWSAzureGCPMachine LearningDeep LearningStatistical Modeling

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Designing Enterprise Ontologies
You must create and maintain semantic models that harmonize data across disparate systems like SAP, Salesforce, and Workday. This requires deep knowledge of knowledge engineering to ensure AI agents have a grounded, accurate understanding of complex business landscapes.
Building Reliable RAG and AI Pipelines
The role focuses on developing production-grade RAG pipelines, vector databases, and embeddings. You are expected to move beyond surface-level AI by grounding generative models in structured enterprise data and business process semantics.
Cross-Functional Stakeholder Alignment
You will act as a bridge between technical AI development and business needs. Success depends on your ability to translate ambiguous requirements from product, engineering, and customer-facing teams into concrete, scalable AI solutions.

Likely challenges to think through

Balancing Semantic Precision with Scalability
Designing ontologies that are both rigorous enough for high-stakes finance processes and scalable across global enterprise systems is difficult. Prepare by discussing how you manage the trade-offs between RDF triple stores and property graph databases in production environments.
Grounding AI in Complex Business Contexts
Generic LLMs often struggle with the specific nuances of enterprise business processes like Procure-to-Pay or Record-to-Report. You will need to demonstrate how you use metadata structures and business process intelligence to prevent hallucinations and ensure accuracy in financial reporting.

Practice your answers

Technical practice

Enterprise Ontology Design

What to coverExplain your approach to designing an enterprise-level ontology that harmonizes data across disparate systems like SAP, Salesforce, and Workday. Discuss how you ensure semantic interoperability and maintain consistency across complex business landscapes.

RDF vs. Property Graph Databases

What to coverCompare and contrast RDF triple stores with property graph databases. Provide a scenario where you would choose one over the other based on requirements for schema flexibility, query complexity, or reasoning capabilities.

RAG Pipeline Optimization

What to coverDescribe how you would build a RAG pipeline to ensure high accuracy for enterprise AI agents. Focus on how you integrate domain-specific knowledge graphs or semantic models to improve retrieval precision beyond standard vector similarity.

Semantic Modeling for Finance and Spend

What to coverHow do you translate complex business processes like Procure-to-Pay or Order-to-Cash into a semantic model? Explain the importance of grounding AI agents in these specific business process semantics.

Graph Query Languages

What to coverDiscuss your experience with SPARQL, Cypher, or GQL. Provide an example of a complex query you have written to extract insights from a knowledge graph and how you optimized its performance.

Agentic AI and Reasoning Frameworks

What to coverExplain your experience with agentic AI architectures, specifically regarding planning, orchestration, and tool use. How do you ensure an AI agent remains reliable when executing multi-step reasoning tasks?

Productionizing AI/ML Solutions

What to coverDescribe your process for deploying AI models into production, including lifecycle management and continuous improvement. How do you handle model evaluation and monitoring when dealing with real-world enterprise datasets?

Cloud Infrastructure for AI

What to coverDiscuss your experience leveraging cloud platforms like Databricks, AWS, Azure, or GCP for scalable AI workflows. How do you manage data pipelines and infrastructure to support large-scale semantic retrieval?

Behavioural practice

Translating Ambiguous Business Challenges

What to coverDescribe a time you had to translate an ambiguous business challenge into a concrete AI solution. How did you engage with cross-functional stakeholders to define the scope and ensure the final output met their needs?

Cross-Functional Collaboration

What to coverProvide an example of working with product, engineering, and business teams to deliver a project. How did you manage conflicting priorities or differing technical perspectives to reach a successful outcome?

Ownership and Accountability

What to coverTell me about a project where you owned an AI solution from concept through deployment. What were the biggest challenges you faced, and how did you ensure the solution remained reliable and trustworthy for the end users?

Continuous Learning in AI

What to coverThe field of GenAI and semantic systems evolves rapidly. Describe a specific instance where you had to quickly learn and apply a new technology or framework to solve a pressing project requirement.

Sharpen your knowledge

Topics to prepare

Enterprise Ontology Design
Focus on the conceptual modeling of business domains to support AI grounding. Exercise: Map a standard business process, such as Procure-to-Pay, into a formal ontology structure. This is directly relevant to the role's requirement for building semantic foundations for SAP AI agents.
Knowledge Graph vs. Property Graph Trade-offs
Study the architectural differences between RDF triple stores and property graph databases. Exercise: Compare the query performance and schema flexibility of SPARQL-based RDF systems versus Cypher-based property graphs for a complex enterprise dataset. This addresses the explicit requirement for knowledge engineering expertise.
Agentic AI and Reasoning Frameworks
Understand the architecture of multi-agent systems, including planning, orchestration, and tool-use capabilities. Exercise: Design a high-level architecture for an AI agent that retrieves data from multiple enterprise sources (e.g., Salesforce and SAP) to answer a complex finance query. This aligns with the preferred qualification for agentic AI architectures.

Skills to revise

Graph Query Languages
Gain proficiency in SPARQL, Cypher, or GQL. Exercise: Write a series of complex queries to extract multi-hop relationships from a sample knowledge graph. This is a core technical requirement for maintaining enterprise semantic models.
RAG and Vector Database Implementation
Develop hands-on experience with building retrieval-augmented generation pipelines. Exercise: Build a RAG pipeline using Python and a vector database that incorporates metadata filtering to ensure high-precision retrieval. This is essential for the role's focus on reliable, production-grade AI solutions.
Semantic Modeling with W3C Standards
Practice using OWL, RDF, and SHACL for data validation and interoperability. Exercise: Define a set of SHACL shapes to validate the integrity of a business data model. This is highly relevant to the preferred qualification for deep expertise in the W3C stack.

Your preparation checklist

Map Enterprise Ontologies to Business Processes
Prepare a technical walkthrough demonstrating how you have designed ontologies to harmonize disparate data sources. Focus on your ability to map structured business data from ERP or CRM systems into a unified semantic layer, specifically referencing your experience with RDF, OWL, or property graph models.
Demonstrate RAG and Knowledge Grounding Expertise
Develop a concise case study of a production-grade RAG pipeline you have built. Explain how you utilized vector databases and embeddings to ground LLM outputs in specific enterprise domain knowledge, ensuring accuracy and reliability for business-critical applications.
Prepare for Technical Discussions on Graph Querying
Review your proficiency in SPARQL, Cypher, or GQL. Be ready to discuss the architectural trade-offs between RDF triple stores and property graph databases when scaling AI solutions for complex enterprise environments.
Align Experience with Finance and Spend Domains
Identify specific examples from your career where you applied machine learning or semantic modeling to financial or procurement processes. Prepare to explain how you translated these domain-specific challenges into scalable AI solutions.
Formulate Strategic Questions for the Data Labs Team
Prepare thoughtful questions regarding the team's current approach to multi-agent architectures, how they handle semantic interoperability across SAP's complex product ecosystem, and the specific challenges they face in maintaining AI reliability for Finance and Spend workflows.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in India. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…