The role involves designing and building deployment infrastructure to move inference code from merge to production. The engineer will focus on orchestrating validation and scheduling deployments while navigating resource constraints related to accelerator chips. The work includes improving capacity-aware scheduling and extending observability tools to track code and validation status. Additionally, the position requires driving down cycle times through pipeline architectures that minimize dependencies. The engineer will also optimize fleet rollout strategies and evolve self-service model onboarding processes.

Candidates must possess at least five years of experience in building deployment or delivery infrastructure at scale. Proficiency in Kubernetes, container orchestration, and managing complex state machines is required. The role necessitates a bachelor’s degree or an equivalent combination of education and professional experience. Strong communication skills are essential for collaborating with cross-functional teams.