Senior Inference Optimization Engineer for Multimodal AI

Rhoda AI; Mountain View (CA); On-site; USD 190,000 - 260,000; Full time

Job summary

Rhoda AI in Mountain View, CA is seeking an Inference Optimization Engineer to accelerate multimodal foundation models across diverse hardware, from AMD Instinct GPUs to embedded Jetson-class compute. You will decide workload mapping and optimize for latency, throughput, and price per inference.

Collaborate with researchers, robotics, and infrastructure teams to port models to non-NVIDIA backends, build portable kernels, and craft cost-performance benchmarks; own end-to-end inference

Qualifications

• 2+ years of experience in inference optimization or ML systems.
• Hands-on PyTorch experience.
• Strong understanding of compute, memory bandwidth, memory capacity, and interconnect bottlenecks.
• Experience with model optimization techniques: quantization, pruning, distillation.
• Familiarity with inference serving frameworks.

Responsibilities

• Own cost-efficient inference end-to-end across vision, language, and control workloads.
• Port and optimize models for non-NVIDIA and lower-cost hardware.
• Own on-robot inference on embedded compute with tight latency/power limits.
• Evaluate hardware with cost/performance models and provide recommendations.
• Build hardware-portable optimizations and kernels across vendors.

Skills

PyTorch

Inference optimization

ML systems

Compiler techniques

Performance debugging

Tools

XLA

TensorRT

IREE/MLIR

TVM

Triton

Job description

Rhoda AI in Mountain View, CA is seeking an Inference Optimization Engineer to accelerate multimodal foundation models across diverse hardware, from AMD Instinct GPUs to embedded Jetson-class compute. You will decide workload mapping and optimize for latency, throughput, and price per inference.

Collaborate with researchers, robotics, and infrastructure teams to port models to non-NVIDIA backends, build portable kernels, and craft cost-performance benchmarks; own end-to-end inference