Job Type: Full-time, Direct W2 Salary Range: $100,000–$150,000 annually Experience Required: 6+ years

Sponsorship: U.S. citizens, green card holders, EAD holders, and H-1B transfer candidates are encouraged to apply. We are unable to sponsor new H-1B petitions for this position.

Key Responsibilities - Design and operate GPU and accelerator infrastructure for training and inference across on-prem clusters, cloud-managed services, and hybrid configurations. - Build scheduling, queuing, and resource-sharing systems to maximize accelerator utilization across multiple teams. - Integrate PyTorch, JAX, DeepSpeed, FSDP, Megatron-LM, and Ray Train into a unified AI platform. - Operate high-performance storage systems and data pipelines to feed training workloads at near-line-rate. - Design networking architectures supporting RDMA, InfiniBand, NCCL, and high-bandwidth communication. - Build observability for AI workloads, including utilization, throughput, training stability, and failure analytics. - Implement checkpointing, restart, and fault-tolerance patterns for long-running traini

Job Type: Full-time, Direct W2 Salary Range: $100,000–$150,000 annually Experience Required: 6+ years

Sponsorship: U.S. citizens, green card holders, EAD holders, and H-1B transfer candidates are encouraged to apply. We are unable to sponsor new H-1B petitions for this position.

Key Responsibilities - Design and operate GPU and accelerator infrastructure for training and inference across on-prem clusters, cloud-managed services, and hybrid configurations. - Build scheduling, queuing, and resource-sharing systems to maximize accelerator utilization across multiple teams. - Integrate PyTorch, JAX, DeepSpeed, FSDP, Megatron-LM, and Ray Train into a unified AI platform. - Operate high-performance storage systems and data pipelines to feed training workloads at near-line-rate. - Design networking architectures supporting RDMA, InfiniBand, NCCL, and high-bandwidth communication. - Build observability for AI workloads, including utilization, throughput, training stability, and failure analytics. - Implement checkpointing, restart, and fault-tolerance patterns for long-running traini