Robotics AI Inference Optimizer — On-Device & Cloud

Digitailor; Mountain View (CA); On-site; USD 140,000 - 210,000; Full time

Job summary

Rhoda AI is building generalist robotic systems and seeking an Inference Optimization engineer to accelerate model inference across diverse hardware. You will optimize vision, language, and sensor models for speed and cost, while balancing on-robot and cloud workloads to meet real-time constraints.

You will collaborate with research, robotics, and infrastructure teams, leveraging PyTorch and compiler toolchains to deliver portable, efficient solutions on AMD ROCm, TPU, and consumer GPUs.

Qualifications

• 2+ years of experience in inference optimization, ML systems, ML compilers, or a closely related field.
• Deep hands-on experience with PyTorch.
• Strong understanding of compute, memory bandwidth, memory capacity, and interconnect bottlenecks.
• Experience with model optimization techniques: quantization (INT8/FP8/FP4), distillation, pruning, and compilation tooling.
• Familiarity with inference serving frameworks and their support for different hardware.

Responsibilities

• Own cost-efficient inference end-to-end across multiple hardware targets.
• Port and optimize models for non-NVIDIA and lower-cost hardware.
• Build hardware-portable optimizations and write kernel-level code where speedups justify them.
• Write clear recommendations on hardware for serving, offline evaluation, and on-robot use.
• Track latency, throughput, and cost per inference across model versions.

Skills

PyTorch

Inference optimization

ML systems

Bottleneck analysis

Ownership mindset

Tools

torch.compile

XLA

TensorRT

IREE/MLIR

TVM

vLLM

Triton Inference Server

Jetson/Thor tooling

Job description

Rhoda AI is building generalist robotic systems and seeking an Inference Optimization engineer to accelerate model inference across diverse hardware. You will optimize vision, language, and sensor models for speed and cost, while balancing on-robot and cloud workloads to meet real-time constraints.

You will collaborate with research, robotics, and infrastructure teams, leveraging PyTorch and compiler toolchains to deliver portable, efficient solutions on AMD ROCm, TPU, and consumer GPUs.