Software Development Engineer III / Sr. Software Development Engineer - AI Platform

Workday Workday · Enterprise · Boulder, CO

The Agent Runtime team builds Workday’s secure execution platform for enterprise AI agents, focusing on enforcing policy through architecture for data access, tool use, model routing, and auditability. The role involves designing and building core runtime capabilities for policy-governed agent execution, including sandbox execution, sidecar interfaces, enforcement flows, and production reliability, with a strong emphasis on observability and integration with other teams.

What you'd actually do

  1. Design and implement core runtime services for secure, isolated agent execution.
  2. Build deterministic enforcement pathways for data, context, tool, and model boundaries.
  3. Define and evolve service interfaces and contracts across maincar and sidecar components.
  4. Improve performance, scalability, and resilience for high-concurrency workloads.
  5. Build deep observability into runtime behavior, policy decisions, and audit traces.

Skills

Required

  • 8+ years experience in software development engineering, architecting, building, and scaling robust and efficient software systems.
  • 6+ years of software engineering experience building backend or platform systems.
  • Strong experience with distributed systems, APIs, and service-oriented architecture.
  • Experience building secure systems with least-privilege patterns and policy controls.
  • Proficiency in at least one systems or backend language (for example, Java, Go, Rust, or Python).
  • Experience operating production services with strong reliability and observability practices.
  • Demonstrated ownership of complex technical projects from design through production.
  • Strong collaboration and communication skills in cross-functional environments.

Nice to have

  • 3+ years of software engineering experience in backend or platform development.
  • Solid foundation in distributed systems fundamentals, APIs, and data modeling.
  • Proficiency in one or more programming languages used for backend systems.
  • Experience building and shipping production services with testing and monitoring.
  • Working knowledge of security fundamentals, including authentication, authorization, and least-privilege concepts.
  • Ability to debug complex issues and deliver reliable fixes in production environments.
  • Strong written and verbal communication skills and effective team collaboration.

What the JD emphasized

  • secure execution platform for enterprise AI agents
  • policy by architecture
  • deterministic controls
  • policy enforcement
  • observability at scale
  • secure, isolated agent execution
  • deterministic enforcement pathways
  • production reliability

Other signals

  • AI agents
  • runtime platform
  • policy enforcement
  • data access
  • tool use
  • model routing
  • observability