Compute Infrastructure Engineer – NeuraGym (human)

Job description preview

From Neura Robotics’s original posting

Your mission & challenges As Compute Infrastructure Engineer – NeuraGym you work at the intersection of server architecture, virtualisation, container orchestration and data processing, with a strong focus on building a scalable compute and storage infrastructure for multimodal robot data. Your core responsibility is designing, building and operating the system landscape in the NeuraGym so that recording and processing on large data volumes run reliably, performantly and reproducibly. You play a decisive role in enabling high-quality embodied AI systems by turning individual servers, workstations and edge devices into a distributed, maintainable and extensible compute system. • Server infrastructure planning: You plan and design the NeuraGym server infrastructure, including sizing of compute, GPU, storage and network, rack layout, power and cooling concept, and procurement in coordination with the team. • Virtualisation: You build and operate virtualisation environments based on hypervisors such as Proxmox and provide development, test and production environments. • Kubernetes & GPU scheduling: You design and operate Kubernetes clusters (or comparable orchestration) for data…