Senior/Staff Software Engineer - Planner Frameworks Pipeline

TL;DR

Senior/Staff Software Engineer - Planner Frameworks Pipeline (Ray/Kubernetes): Build and optimize large-scale simulation and validation pipelines for a self-driving vehicle, with an accent on cost efficiency, reliability, and performance. Focus on designing production-grade APIs/SDKs and improving ML training pipelines that support autonomous behavior development.

Location: Foster City, CA / San Francisco, CA

Company

Zoox develops autonomy software for a self-driving vehicle.

What you will do

• Improve cost efficiency, reliability, and performance of validation and simulation pipelines.
• Create production-grade APIs, SDKs, and tools to enable a wide range of autonomous behavior validations.
• Improve ML training pipelines supporting the autonomous behavior organization.
• Contribute to framework architecture, reliability, and efficiency, and expand capabilities for new validation use cases.
• Work closely with engineers developing on-vehicle algorithms and models to support end-to-end validation use cases.

Requirements

• Bachelor’s degree in Computer Science or related field and 8+ years of industry experience.
• Experience optimizing large-scale distributed systems for cost and efficiency.
• Experience with AWS or similar cloud providers.
• Proficiency with Python and familiarity with C++.
• Track record of keeping production systems running with high availability.

Nice to have

• Exposure to ML workloads (training, inference, data generation) with a cost-optimization perspective.
• Background in algorithmic optimization or performance investigation.
• Experience with Ray, especially Ray Core and Ray Data.
• Understanding of the robotic data lifecycle; robotics/autonomous systems experience not required.

Culture & Benefits

• Hybrid workplace based in Foster City, CA / San Francisco, CA.
• Hands-on role with high independence and ownership.
• Small, scrappy team within the larger Autonomy organization.
• Close collaboration with autonomy engineers to focus on end-to-end validation use cases.