remote
GPU Systems Engineer - DeepInfra
Systems Engineer
GPU Systems Engineer to design and optimize high-performance computing systems leveraging GPU acceleration for AI workloads.
About the role
Key Responsibilities
- Design and optimize GPU-accelerated systems for AI workloads
- Develop low-level drivers and runtime systems for GPUs
- Collaborate with hardware vendors to influence product roadmaps
- Profile and optimize memory access patterns for maximum throughput
- Implement power management and thermal solutions for GPU clusters
- Create tools for GPU utilization monitoring and debugging
Requirements
- 5+ years in systems programming or GPU computing
- Expertise in CUDA, OpenCL, or similar GPU programming models
- Strong understanding of computer architecture and GPU microarchitecture
- Experience with Linux kernel development and drivers
- Knowledge of parallel computing and memory hierarchies
Skills
gpu architecturecudaopenclsystem designperformance tuninglinux kernel