GPU Infrastructure Specialist
- Niagara-on-the-Lake, ON, Canada
Full-time
Applications open in a new tab.
Job details
- Job type
- Full-time
- Apply by
- Posting language
- English
- Posted
Original job description
TAIRC seeks a GPU Infrastructure Specialist to support GPUs and hardware accelerators used for AI workloads. The role manages GPU clusters, optimizes GPU utilization, advises researchers on hardware requirements, and ensures hardware reliability.
Responsibilities
Responsibilities
- Manage GPU clusters: Set up, configure, and maintain GPU-based servers
- Optimize GPU utilization: Implement strategies for multi-GPU training, memory management, and load balancing
- Collaborate with researchers: Provide advice on hardware requirements for large-scale experiments
- Ensure reliability: Monitor hardware health and replace components as needed
More about this job
Responsibilities
Manage and maintain GPU-based servers and clusters, optimizing utilization through multi-GPU training strategies, memory management, and load balancing. Advise researchers on hardware needs for large-scale experiments and monitor hardware health, replacing components as needed.
Requirements
The role calls for experience managing GPU clusters and supporting hardware accelerators for AI workloads. It also requires the ability to optimize GPU utilization, advise researchers on hardware requirements, and maintain hardware reliability.
Skills
- GPU Cluster Management
- GPU Server Configuration
- GPU Utilization Optimization
- Multi-GPU Training
- Memory Management
- Load Balancing
- Hardware Requirements Advising
- Hardware Health Monitoring
- Hardware Reliability
- Component Replacement
Visa sponsorship
Not detected in the job text
Categories
- Technology
- Engineering
- Science & Research
Keywords
- GPU
- Hardware Accelerators
- AI Workloads
- GPU Clusters
- GPU-Based Servers
- GPU Utilization
- Multi-GPU Training
- Memory Management
- Load Balancing
- Large-Scale Experiments
- Hardware Requirements
- Hardware Health
- Hardware Reliability
- Component Replacement