Staff ML Infra Engineer for Autonomy Platform

Rivian; Sacramento (CA); On-site; USD 207,000 - 258,000; Full time

Job summary

Rivian is seeking a Staff Software Engineer for its Autonomy ML Infrastructure team. You will design and build a training platform for autonomous driving models, including the control plane, storage/I-O, observability, and the training framework.

You will lead one or two areas end-to-end and collaborate with model training, cloud, and data teams to raise overall goodput. The role emphasizes technical direction, architectural leadership, and practical impact at scale.

Qualifications

• 5+ years of software engineering with distributed-systems in production.
• Staff-level technical leadership with strategy and prioritization responsibilities.
• Hands-on experience building or operating large-scale compute or ML infrastructure.
• Deep expertise in one of: cluster scheduling and multi-tenancy, training data storage and I/O, observability and performance engineering, or distributed training frameworks.

Responsibilities

• Evolve the control plane to be the single entry point for accelerator clusters across providers and silicon types.
• Design scheduling to maximize fleet goodput with gang scheduling, multi-factor priorities, and fair sharing.
• Build monitoring and profiling tools across layers to provide a unified view for platform and users.

Skills

Python

Go

Rust

C++

Education

Bachelors in CS or related field

Tools

Kubernetes

PyTorch Distributed

Ray

JAX

Job description

Rivian is seeking a Staff Software Engineer for its Autonomy ML Infrastructure team. You will design and build a training platform for autonomous driving models, including the control plane, storage/I-O, observability, and the training framework.

You will lead one or two areas end-to-end and collaborate with model training, cloud, and data teams to raise overall goodput. The role emphasizes technical direction, architectural leadership, and practical impact at scale.