About this role
TensorWave is seeking an ML Performance Engineer to join our MLOps team. You will validate new AI compute clusters by running large-scale training and inference workloads, diagnose performance issues, and produce benchmarks. This role offers a unique opportunity to directly influence the performance and reliability of our cutting-edge AI infrastructure.
What we are looking for 6
- Validate new GPU cluster deployments with large-scale ML workloads
- Diagnose and resolve ML performance and reliability issues
- Provide feedback to improve managed Kubernetes and Slurm platforms
- Benchmark ML model training and inference performance
- Build tooling for validation and benchmarking workflows
- Experience with distributed training frameworks (PyTorch FSDP, Megatron-LM, DeepSpeed)
Work authorization
- Must be authorized to work without sponsorship.