Staff Software Engineer, Compute & Storage, Autonomy
Core
Own the distributed compute and storage platform that powers every autonomy workload, enabling fast data access and efficient GPU training for a fleet of 100,000+ vehicles.
Role type
Staff Software Engineer (Platform Infrastructure)
Builds
Distributed compute orchestration, petabyte-scale object storage, and multi-cloud data paths for AI training and simulation.
Domain
Autonomous vehicles / AI Platform / Cloud Infrastructure
Deliverable
production ML models
Required skills
Kubernetes orchestration, large-scale object storage, distributed systems design, Go/C++/Rust, Python, infrastructure as code, queueing systems, observability, cost optimization
Preferred skills
GPU cluster management, autonomous vehicle domain, multi-cloud operations, Linux internals
Technologies
Kubernetes, Ray, Spark, Flink, Dask, Go, C++, Rust, Python, Terraform, AWS CDK, CloudFormation, SQS, Kafka, Kinesis, Prometheus, Grafana, Datadog, CloudWatch, AWS
Responsibilities
Own architecture and roadmap for distributed compute platform including scheduling, autoscaling, and spot/preemption strategies. Build scalable tools and APIs to execute high-level job requests for non-infrastructure engineers. Profile and optimize cluster efficiency to eliminate GPU fragmentation and reclaim stranded capacity. Design petabyte-scale storage architecture with tiering, lifecycle policies, and caching layers. Manage multi-cloud compute and data paths including replication and cross-provider egress economics. Drive throughput for heavy workloads like training data loading and batch resimulation. Define SLAs for job admission and data availability. Own platform observability, metrics, dashboards, and alerting for job throughput and cost. Set technical standards and mentor senior engineers through design and code reviews.
Seniority
Staff, technical leadership & deep hands-on IC