Machine Learning Performance Engineer, Training
Tower Research Capital
- Salary
- $200k
- Experience
- 3+ yrs
- Location
- New York City, New York, United States
- Job type
- Full-time
Required skills
- PyTorch
- JAX
- Python
- C++
- CUDA
- Triton
- CUTLASS
- cuBLAS
- cuDNN
- NCCL
- FSDP
- DeepSpeed
- Megatron-LM
- XLA
- Nsight Systems
- Nsight Compute
- PyTorch Profiler
- InfiniBand
- RDMA
- NVLink
- NVSwitch
- Kubernetes
- Slurm
- Ray
About the role
3+ years optimizing ML training workloads; deep knowledge of PyTorch or JAX; strong Python and C++; GPU kernel development with CUDA, Triton, CUTLASS, cuBLAS, cuDNN; distributed training with NCCL, FSDP, DeepSpeed, Megatron-LM, XLA; performance tools like Nsight Systems, Nsight Compute, PyTorch Profiler.
About Tower Research Capital
Proprietary quantitative trading firm employing traders, engineers, researchers, and business-support staff to trade global financial markets.
This page is fully interactive when JavaScript is enabled. Please enable JavaScript to apply or browse related roles.
