Staff Software Engineer, ML Infra, Autonomy
Core
Design and build the ML infrastructure platform (control plane, storage/I/O, observability, training framework) that trains and evaluates Rivian's autonomous driving models at fleet scale.
Role type
Staff Software Engineer, ML Infrastructure
Builds
Training platform for hundreds of ML engineers using thousands of GPUs on Kubernetes over petabyte-scale sensor data
Domain
Autonomous driving, Machine Learning Infrastructure, Distributed Systems
Deliverable
production ML models
Required skills
Distributed systems design, Cluster scheduling and multi-tenancy, Training data storage and I/O, Distributed training frameworks, Observability and performance engineering, Python, Go/Rust/C++, Cloud provider operations, Kubernetes
Preferred skills
Operating systems internals, Large-model training optimization, Ray/KubeRay internals, GPU profiling tools
Technologies
Kubernetes, Kueue, Volcano, Slurm, YARN, PyTorch, Ray Train, JAX, AWS, NCCL
Responsibilities
Design gang scheduling and multi-factor priority for accelerator clusters; Architect storage tiering and shard formats to minimize input wait; Build monitoring and profiling toolchain for distributed workloads; Develop thin training framework layer over open-source libraries; Define platform roadmap and mentor engineers
Seniority
Staff, technical leadership & hands-on IC