Ebay AI Infrastructure DevOPS Engineer 

Netherlands, North Holland, Amsterdam 

999576035

• Architect and design of high-performance storage system for GPU cluster, supporting large checkpoints and low-latency context preemption and reloads.

• Develop monitoring and observability tools for GPU clusters

• Maintain high availability, fault tolerance, and disaster recovery strategies for AI infrastructure

• Work closely with AI/ML engineers, data scientists, and DevOps teams to streamline AI workflows.

What you will bring:

• Masters or PhD in EE or CS

• Over 5 years of experience building HPC systems

• C/C++ Programming – for performance-critical components and integration tasks. Lustre (Paralell filesystems is in C)

• Linux Kernel and OS internals – to optimize system behavior and support kernel-level customization for filesystems and networking

• Filesystems knowledge – with a strong preference for experience in Lustre or similar distributed filesystems

• Kubernetes – for container orchestration and management at scale

• Hardware and Networking familiarity – to work effectively with low-level infrastructure and tuning

Good to have:

• Strong understanding of RDMA, RoCE V2 protocols

• Hands-on experience with GPUs

• Understanding of AI Workflows, training, inferencing

• Understanding of AI/ML Python frameworks (TensorFlow, PyTorch)