ML Systems Engineer, Large-Scale Model Training & RL Infrastructure
$195,200–$262,200 year
On-sitePalo Alto, California, United States
Job Summary
Build and maintain distributed training infrastructure for SFT, continued pretraining, preference optimization, and RL workloads. Integrate frameworks like Megatron-LM, DeepSpeed, PyTorch FSDP/DTensor, Ray, verl, and OpenRLHF while implementing parallelism strategies for tensor, pipeline, and data parallelism. Create reliable rollout, reward model serving, replay buffers, checkpointing, and experiment orchestration components. Profile GPU utilization, communication efficiency, and memory usage to improve training throughput. Diagnose failures across NCCL, CUDA, PyTorch, Ray, schedulers, storage, and networking layers. Partner with research scientists to turn algorithmic recipes into scalable, debuggable systems and produce design docs, incident reports, and operational tooling.
Required Qualifications
- Strong Python and PyTorch engineering skills
- Hands-on experience with distributed model training, large-scale ML systems, or GPU cluster workloads
- Practical understanding of transformer training bottlenecks, memory pressure, gradient/optimizer state, communication overhead, and checkpointing
- Experience debugging production or research training jobs across multiple GPUs or nodes
- Ability to reason quantitatively about throughput, utilization, memory, reliability, cost, and research velocity
- Strong communication skills and ability to collaborate with researchers, ML engineers, platform engineers, and leadership
- Applicant must be authorized to work in the country in which they apply
Desired Qualifications
- Experience with Megatron-LM, DeepSpeed, PyTorch FSDP/DTensor, Ray, Slurm, Kubernetes, or large internal training platforms
- Experience with RL infrastructure frameworks such as verl, slime, AReaL, OpenRLHF, TRL, or custom PPO/GRPO/RLHF systems
- Familiarity with NCCL, CUDA, Triton, Nsight, InfiniBand, RDMA, RoCE, H100/H200/B200 clusters, or storage/network bottlenecks
- Experience supporting SFT, DPO, PPO, GRPO, RLAIF, reward model serving, rollout generation, or agent training workloads
- Open-source contributions to distributed training, RL infrastructure, PyTorch, Ray, Megatron, DeepSpeed, or related systems
Hiring someone like this?
Get your role in front of qualified candidates on Sorce.