Senior Research Engineer, Robotics & AI Foundations

Amazon; San Francisco (CA); On-site; USD 180,000 - 320,000; Full time

Job summary

Frontier AI & Robotics is hiring a Senior Research Engineer embedded in our science team to transform research into high-performance production systems. You will collaborate with scientists to build and optimize large-scale transformer models for robotics applications, balancing deep technical work with model architecture decisions.

You will leverage PyTorch, CUDA, and NVIDIA tools to meet ambitious performance targets, pushing the boundaries of robotics intelligence from perception to

Qualifications

• 5+ years of professional software development or a degree in CS as alternate.
• 5+ years of programming in at least one language.
• 5+ years in design or architecture of large systems.
• Experience mentoring, tech lead, or engineering leadership.
• Expertise in Python, PyTorch, and CUDA programming.
• Experience with TensorRT or ML optimization frameworks.
• Ability to optimize ML models for production.
• Experience with research teams and distributed training for foundation models.

Responsibilities

• Design, implement, and optimize distributed training systems across GPUs/nodes.
• Develop high-performance optimizations to maximize throughput and efficiency.
• Create reusable frameworks to improve training reproducibility and scalability.
• Establish reliability, maintainability, and security standards for rapid iteration.
• Collaborate with researchers to influence model architectures for hardware utilization.
• Develop benchmarking frameworks to measure and optimize model performance.
• Prototype acceleration approaches using new compilation techniques.

Skills

Python

PyTorch

CUDA programming

Mentor/tech lead

Distributed training

Research collaboration

Model optimization

Education

Bachelor's degree in computer science or equivalent

Tools

TensorRT

ONNX Runtime

CUDA

Megatron

TorchTitan

Job description

Frontier AI & Robotics is hiring a Senior Research Engineer embedded in our science team to transform research into high-performance production systems. You will collaborate with scientists to build and optimize large-scale transformer models for robotics applications, balancing deep technical work with model architecture decisions.

You will leverage PyTorch, CUDA, and NVIDIA tools to meet ambitious performance targets, pushing the boundaries of robotics intelligence from perception to