Robotics Inference Engineer Intern — Production-Ready

Rhoda; Mountain View (CA); On-site; USD 17,000 - 23,000; Full time

Benefits offered by this job

Mentorship from experienced engineers

Access to modern GPU compute

Job summary

Rhoda AI is building generalist robots with a full-stack approach, including high-performance hardware and foundation world models. This internship focuses on accelerating model inference across the stack, from data pipelines to GPU kernels, and owning a high-impact project from start to production.

You will profile, optimize, and ship improvements in latency, throughput, and memory, collaborating with infra and research engineers.

Qualifications

• Pursuing BS/MS/PhD in CS/EE or related field.
• Strong Python and/or C++ programming skills.
• Solid understanding of deep learning fundamentals and PyTorch.
• Ability to read performance data and reason about latency/trade-offs.

Responsibilities

• Profile and benchmark inference for large models to find bottlenecks in latency, throughput, memory.
• Apply and evaluate inference optimization techniques like quantization and low‑precision inference.
• Optimize data movement: video decoding, input pipelines, host-device transfer, caching, I/O paths.

Skills

Python/C++ programming

Deep learning fundamentals

Performance data analysis

Latency and throughput trade-offs

Education

BS/MS/PhD in CS/EE or related

Tools

PyTorch

TensorRT

ONNX Runtime

TVM

CUDA

Triton

CUTLASS

Job description

Rhoda AI is building generalist robots with a full-stack approach, including high-performance hardware and foundation world models. This internship focuses on accelerating model inference across the stack, from data pipelines to GPU kernels, and owning a high-impact project from start to production.

You will profile, optimize, and ship improvements in latency, throughput, and memory, collaborating with infra and research engineers.