Staff ML Infra Engineer — Training & Inference (Autonomy)

Rivian Automotive; England; On-site; GBP 140,000 - 200,000; Full time

Job summary

Rivian is seeking a Staff Engineer to join the Perception team, focusing on state-of-the-art ML infrastructure for training and inference of large autonomous driving models. You will help establish scalable systems and leverage advanced ML techniques to improve safety-critical features across Rivian's vehicles.

You will design and optimize training pipelines, accelerate transformer-based models, and collaborate across teams to set technical standards and best practices.

Qualifications

• PhD in CS/CE/EE or equivalent industry experience.
• Extensive knowledge of PyTorch, TensorFlow or JAX.
• Experience with large-scale distributed training.
• Proven ability to profile and optimize training and inference performance.

Responsibilities

• Optimize performance of Deep Learning training workloads on NVIDIA GPU systems at large scale.
• Reduce latency of model inference and pre/post-processing on onboard systems.
• Design, train, and deploy large deep learning models using large datasets.
• Define cross-team technical strategies and elevate engineering practices.
• Benchmark and adopt modern tools, frameworks, and architectural patterns.

Skills

PyTorch

TensorFlow/JAX

Training frameworks

Transformer models

Distributed training

Model profiling

Education

PhD in CS/CE/EE

Job description

Rivian is seeking a Staff Engineer to join the Perception team, focusing on state-of-the-art ML infrastructure for training and inference of large autonomous driving models. You will help establish scalable systems and leverage advanced ML techniques to improve safety-critical features across Rivian's vehicles.

You will design and optimize training pipelines, accelerate transformer-based models, and collaborate across teams to set technical standards and best practices.