Overview
The AI Platform Engineers will be responsible for the engineering and development of an enterprise AI infrastructure platform, primarily focusing on Red Hat OpenShift AI environments and NVIDIA GPU infrastructure. This hybrid role will engage with complex technologies to ensure optimal performance of distributed GPU workloads and accelerated networking, requiring collaboration with technology vendors for troubleshooting and support. The initial contract spans six months, with the possibility of extensions based on project needs.
Responsibilities
- Engineer and support Red Hat OpenShift AI and AI/ML workloads.
- Manage NVIDIA GPU infrastructure including drivers and CUDA integration.
- Monitor and allocate GPU resources, ensuring efficient scheduling.
- Implement high-performance networking technologies relevant to AI workloads.
- Troubleshoot issues across OpenShift, GPUs, and physical infrastructure.
- Collaborate with vendors to resolve complex platform issues.
- Perform performance and resilience testing for AI platforms.
Requirements
- Proven experience with Red Hat OpenShift and AI/ML workloads.
- Knowledge of NVIDIA GPU infrastructure and resource management.
- Familiarity with GPU scheduling, monitoring, and utilisation tools.
- Experience with high-performance networking technologies like NVIDIA Spectrum-X.
- Understanding of multi-node GPU communication and NCCL testing.
- Ability to troubleshoot problems in complex infrastructure setups.
- Experience working with technology vendors to resolve issues.