Leads the team this role joins
Runs the team day to day
Decides on early hires
From your own Gmail and LinkedIn.
About the role
This is a senior lead MLOps engineering role at NVIDIA in Santa Clara, California, for an engineer who supports large-scale GPU and HPC systems used for AI research. The role works one to one with researchers on LLM, autonomous-vehicle and generative workloads, diagnosing training and inference performance problems across compute, network and storage. The posting describes the position as remote, though the location listed is Santa Clara, CA.
What you’ll do
- Set the technical direction for the MLOps team and serve as the senior hands-on engineer for complex customer and performance issues.
- Work with researchers to diagnose training and inference performance problems.
- Debug large-scale distributed training, including NCCL collectives, RDMA, parallel storage, containers, Slurm and Kubernetes.
- Find recurring problems across customers and work with platform and partner teams to fix them at scale.
- Test the platform by running the workloads customers run.
- Produce training, guidance and reference tests, and share learnings publicly.
- Mentor engineers, run design reviews and keep decisions documented.
What they’re looking for
- BS or MS in Computer Science, Engineering or a related field, or equivalent experience.
- 8+ years operating large-scale GPU or HPC infrastructure, including distributed training and inference.
- Hands-on experience with Slurm or Kubernetes and with containers such as Docker, Enroot, Apptainer or Singularity.
- Understanding of a distributed training framework such as PyTorch, NeMo, Megatron or JAX, and an inference server such as Triton, TorchServe or Ray Serve.
- Experience profiling distributed training with the PyTorch profiler, Nsight Systems or Nsight Compute.
- Understanding of NCCL collectives such as all-gather and all-to-all.
- Fluent in Python and at least one of Go, Rust or C++.
- Ability to isolate performance problems to compute, network or storage.
- Strong written and verbal communication.
Nice to have
- Leading bring-up or acceptance testing of GPU clusters with thousands of nodes, including MLPerf benchmarking.
- Deep knowledge of InfiniBand, RDMA, Lustre or GPFS.
- CUDA programming background.
- Observability experience with OpenTelemetry, Grafana or ClickHouse.
- Technical leadership and vendor engagement experience.
What’s on offer
- Base salary of USD 187,500 to 364,000 for Level 4, or USD 232,500 to 442,000 for Level 5, based on location and experience.
- Equity and a comprehensive benefits package.
- Applications accepted at least until 23 November 2026.
Skills
Questions about this role
Is this role remote?
The posting lists the location as Santa Clara, CA, but its metadata marks the role as remote.
How much does it pay?
The posted base salary is USD 187,500 to 364,000 for Level 4, or USD 232,500 to 442,000 for Level 5.
How many years of experience do I need?
The posting asks for 8 or more years designing, operating or supporting large-scale GPU or HPC infrastructure.
Until when can I apply?
Applications will be accepted at least until November 23, 2026.