ML Training Infrastructure Engineer (AI)

Location: Redwood City, California, United States; on-site

Salary: $220,000–$320,000 base salary per year, plus equity

Company

Builds general-purpose robots powered by an embodied AI foundation model and deployed across multiple commercial industries.

What you will do

• Architect and own large-scale, multi-cloud GPU training infrastructure for massive multimodal models.
• Implement distributed training techniques including sharding, activation checkpointing, mixed precision, and memory optimization with ZeRO and FSDP.
• Build research codebases and Kubernetes/SLURM scheduling systems for fast iteration, automated retries, and failure recovery.
• Design high-throughput pipelines for terabytes of multimodal robot data, including video, proprioception, and 3D signals.
• Develop low-latency inference pipelines for real-time robot control using quantization, distillation, TensorRT, and Triton.
• Profile GPU utilization, I/O bottlenecks, memory fragmentation, and inter-node communication across the compute fleet.

Requirements

• 7+ years of engineering experience, including leadership of technical projects in HPC or ML infrastructure.
• Deep experience with PyTorch and distributed training frameworks such as DeepSpeed and Accelerate.
• Hands-on experience managing cloud GPU environments on GCP or AWS and using Kubernetes.
• Understanding of distributed systems, race conditions, memory management, NCCL, and inter-node communication.
• Ability to design, build, and operate infrastructure end to end.
• Availability to work on-site in Redwood City, California, United States.

Nice to have

• Experience with robotics data formats such as MCAP and Protobuf or multimodal models such as VLAs.
• Experience with custom Triton kernels, compilers, or runtime optimization.
• Experience as a founding or early-stage infrastructure hire.

Culture & Benefits

• Work on robotics technology designed for real-world commercial applications.
• Collaborate with experienced researchers and engineers from major technology companies.
• Receive equity in addition to the base salary.
• Work in an environment emphasizing technical rigor, mutual respect, problem-solving, and grit.