Machine Learning Performance Engineer
Jane Streetabout 3 hours ago
Responsibilities
- Optimize the performance of machine learning models for training and inference.
- Enhance large-scale training efficiency and low-latency inference in real-time systems.
- Utilize a whole-systems approach to improve storage, networking, and GPU-level performance.
- Debug and optimize training runs' performance end to end.
- Evaluate throughput and goodput at the lowest system levels.
Requirements
- Experience in low-level systems programming and optimization.
- Understanding of modern machine learning techniques and toolsets.
- Low-level GPU knowledge including PTX, SASS, and Tensor Cores.
- Experience with debugging and optimization tools like CUDA GDB and NSight.
- Familiarity with libraries such as Triton, cuDNN, and cuBLAS.
- Background in networking technologies like Infiniband and NVLink.