Member of Technical Staff - GPU Infrastructure
Prime Intellect · Remote · San Francisco · Remote · Asia ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Fully remote
- GPU Infrastructure
- Kubernetes + SLURM
- Customer-facing
- $150k-$300k
Prime Intellect builds the open superintelligence stack, making frontier AI infrastructure available to ambitious teams. As a Member of Technical Staff, you will partner with clients to design optimal GPU cluster architectures and handle deployment strategies for LLM training and inference.
You will serve as the primary technical escalation point, diagnosing complex issues across hardware, networking, and software stacks. The role requires deep expertise in HPC environments, Kubernetes, and NVIDIA GPU ecosystems to ensure high-performance operations for customers.
What you'll do
- Design optimal GPU cluster architectures for client workloads
- Create technical proposals for clusters ranging from 100 to 10,000+ GPUs
- Deploy orchestration systems like SLURM and Kubernetes
- Implement high-performance networking with InfiniBand and RoCE
- Optimize GPU utilization, memory management, and inter-node communication
- Serve as primary technical escalation point for customer issues
- Provide 24/7 on-call support for critical customer deployments
What you'll need
- 3+ years hands-on experience with GPU clusters and HPC environments
- Deep expertise with SLURM and Kubernetes in production settings
- Proven experience with InfiniBand configuration and troubleshooting
- Strong understanding of NVIDIA GPU architecture and CUDA ecosystem
- Experience with infrastructure automation tools like Ansible or Terraform
- Proficiency in Python, Bash, and systems programming
- Track record of customer-facing technical leadership
Nice to have
- Experience with 1000+ GPU deployments
- NVIDIA DGX, HGX, or SuperPOD certification
- Knowledge of distributed training frameworks like PyTorch FSDP
- Experience with AMD MI300 or Intel Gaudi accelerators
- Contributions to open-source HPC or AI infrastructure projects
Stack
- SLURM
- Kubernetes
- InfiniBand
- CUDA
- Python
- Bash
- Ansible
- Terraform
- Docker
- Lustre
How you'll work
- Fully remote position
- 24/7 on-call support required for critical deployments
What you get
- Cash compensation range of $150,000 to $300,000
- Equity incentives included