Senior Software Engineer - Context and Semantics (AI)

Workday•Pleasanton, CA

8d•Hybrid

About The Position

The Data Cloud organization is at the forefront of Workday’s next major wave of innovation, driven by a bold mandate to build the industry’s first true Platform for Agents. We are a high-energy, ambitious team that balances Workday's deep foundational expertise with fresh, forward-thinking perspectives to accelerate our pace of delivery and unlock massive new commercial value. By leveraging Workday’s unique business context, trusted data, and identity guardrails, we are building the critical engine that will power the future of AI and agentic workflows across the enterprise. Within this organization, the Context and Semantics Engineering team acts as the brain trust responsible for making our AI "enterprise-aware." We build the semantic layers, knowledge graphs, query federation engines, and dynamic context systems that ground generative AI in real-world HR and Finance data. By bridging the gap between raw enterprise data and large language models (LLMs), our team ensures that Workday’s autonomous agents can reason accurately, securely, and with deep contextual awareness at a global scale. As a Senior Software Development Engineer in Context and Semantics, you will be a lead builder and technical anchor for the infrastructure powering Workday’s AI Agents. You will be hands-on in designing, coding, and scaling the backend systems that connect our foundational models to highly complex enterprise data. Your day-to-day will sit at the intersection of distributed data engineering and applied AI. You will build advanced RAG (Retrieval-Augmented Generation) pipelines, implement vector search infrastructure, and develop query federation layers (e.g., Trino/LDQ) that seamlessly route LLM data requests across our transactional (OMS) and analytical (Data Lake) stores. We are looking for a highly autonomous, product-minded engineer who thrives in ambiguity, writes exceptionally clean code, and is passionate about solving the unique latency, cost, and security constraints of enterprise AI.

Requirements

8+ years of professional software engineering experience focusing on backend architecture, data engineering, or distributed systems.
5+ years of strong hands-on programming expertise in Java, Python, Scala, or Go.
1+ years of dedicated experience building production-grade LLM applications, RAG pipelines, Vector Search systems, or Knowledge Graphs.

Nice To Haves

Bachelor’s degree (Master’s preferred) in Computer Science, Software Engineering, or equivalent technical field.
Experience building or optimizing distributed SQL query engines (e.g., Trino, Presto, Spark) and bridging operational and analytical data stores.
Deep understanding of modern AI orchestration frameworks (LangChain, LlamaIndex, Semantic Kernel) and embedding models.
Proven ability to architect scalable, high-performance REST/gRPC APIs and microservices.
Experience handling sensitive, highly regulated data with a strong understanding of enterprise security boundaries and data governance.
A highly autonomous builder capable of taking open-ended AI product goals and breaking them down into concrete, scalable engineering realities.
Experience with the Workday platform or enterprise SaaS architectures is a strong plus.

Responsibilities

Architect, build, and deploy robust backend services, APIs, and data pipelines that form the core semantic and context layer for Workday’s AI Agents.
Design and optimize enterprise-grade RAG pipelines, including embedding generation, intelligent chunking strategies, and vector database integration (e.g., Pinecone, Milvus, Elasticsearch).
Build and scale distributed query execution layers (leveraging technologies like Trino) to seamlessly fetch and translate context across transactional systems and data lakes.
Develop dynamic context and memory management workflows using modern AI orchestration frameworks (e.g., LlamaIndex, LangChain, GraphRAG) to enable complex, multi-turn agentic reasoning.
Embed rigorous data privacy, tenant isolation, and Role-Based Access Control (RBAC) directly into the retrieval and context layers to ensure governed AI execution.
Optimize system architectures for low-latency retrieval and high-throughput LLM interactions, tackling constraints around token management and context-window efficiency.
Mentor mid-level and junior engineers, lead code and design reviews, and establish engineering best practices for this new 0-to-1 product area.