Senior AI Engineer - Agent Factory

Workday•Boulder, CO

2d•Hybrid

About The Position

Agent Factory is where Workday’s next chapter gets built. We’re forming small, senior, cross-functional AI teams that bring together product leaders, AI engineers, and full-stack builders to create intelligent agents used by millions of people every day. This is production-grade AI—deeply embedded into Workday’s platform—not research experiments or maintenance work. Teams own problems end to end, collaborate tightly across disciplines, and use the right tools to solve real customer challenges at global scale. You’ll work at the intersection of AI, platform architecture, and human workflows, with the autonomy to shape how agents reason, act, and scale responsibly. High trust, high expectations, and real impact. Engineering, but brighter.

Requirements

8+ years of professional software engineering experience with strong expertise in backend architecture, distributed systems, and API design, plus 1+ years of dedicated focus building production-grade LLM/agentic systems OR 5+ years of experience specifically within Machine Learning Engineering or AI application development, with 2+ years dedicated to shipping LLM-backed products.
2+ years of hands-on experience integrating large models (LLMs, Foundation Models) and modern AI APIs into user-facing enterprise products.
1+ years of experience designing and scaling AI orchestration architectures—including multi-agent frameworks, routing layers, or advanced RAG pipelines.
4+ years of experience optimizing application performance (specifically tackling constraints like API latency and user interaction design), with 1+ years applied to modern LLM constraints (such as token management, cost optimization, and context-window efficiency).
4+ years of proven experience leveraging cloud computing platforms (e.g., AWS, GCP) to deploy highly responsive, scalable systems.
Strong understanding of how to execute governance, guardrails, security layers, and evaluation mechanisms necessary when deploying autonomous agents over sensitive enterprise HR and financial data.
Proven track record of technically leading engineering workstreams within a pod, taking ownership of the development lifecycle, and mentoring junior-to-mid level engineers.
Deep focus on business value, user experience, and applying deep learning/large models directly to solve practical end-user challenges.
Proven ability to architect robust application layers that wrap around AI models, establishing reusable patterns for system predictability, error handling, and seamless UX integration.
Skilled in rapid prototyping, benchmarking model outputs against product requirements, and setting up automated evaluation metrics (e.g., assessing retrieval quality and agentic behavior).
Highly autonomous builder capable of taking open-ended product goals and breaking them down into concrete, scalable engineering realities.

Nice To Haves

Master’s degree in Computer Science, Software Engineering, or equivalent technical field.

Responsibilities

Drive the end-to-end system design, implementation, and product integration for a core domain of Workday’s next generation of intelligent agents.
Connect the brain to the product, focusing on intelligence orchestration and product delivery.
Develop and integrate foundational models safely and reliably into functional, production-grade software.
Design, experiment with, and orchestrate complex agentic workflows.
Translate cutting-edge AI capabilities into scalable business value.
Implement strict guardrails for data privacy, predictability, and explainability within your pod.
Balance domain-level system architecture with rigorous execution.
Solve critical product constraints like latency, cost, and reliability.
Technically lead engineering workstreams within a pod, taking ownership of the development lifecycle.
Mentor junior-to-mid level engineers.
Architect robust application layers that wrap around AI models, establishing reusable patterns for system predictability, error handling, and seamless UX integration.
Rapidly prototype, benchmark model outputs against product requirements, and set up automated evaluation metrics.
Take open-ended product goals and break them down into concrete, scalable engineering realities.