2 months ago
South San Francisco, CA, USASenior / Staff+
H1B sponsor
Base Salary
$200k - $280k/yr
Responsibilities
- Implement, train, and evaluate real-time 3D perception models using multiple cameras and timesteps.
- Deploy perception models onboard resource-constrained computers and optimize their compute and memory usage.
- Build visualization, introspection, evaluation, and data-flywheel tooling for understanding model performance and finding long-tail samples.
- Work with motion planning to define compact subsystem interfaces and track metrics for overall system improvement.
- Ship new production features or models approximately every six weeks.
- At the Senior level, lead architectural decisions, experimentation, and ownership of a model head or backbone.
- At the Staff level, roadmap onboard models and manage cross-functional interfaces.
- Track research developments and apply new approaches to real-time, compute-constrained 3D perception.
Requirements
- At least 5 years of experience for Senior candidates or 8 years for Staff candidates building and deploying deep-learning-based perception systems.
- Experience with 3D geometry, semantic segmentation, or learned multi-view stereo.
- Strong knowledge of classical computer vision, including camera calibration, epipolar geometry, structure-from-motion, and SGBM stereo.
- Expertise in robotics fundamentals, including reference frames, matrix math, SE(3) manifolds, and probabilistic sensor fusion.
- Hands-on experience training, iterating on, and optimizing CNN and transformer architectures on target hardware such as NVIDIA Jetson-class systems.
- An engineering focus on production outcomes, prioritization, and scalable architecture.
- Senior candidates should have experience across the ML lifecycle, including training, data annotation, evaluation, and hard mining.
- Staff candidates should have experience influencing system-wide design across data pipelines, training infrastructure, and compute requirements.
- Experience applying machine learning to hardware applications is preferred.
Benefits
- The compensation package may include equity compensation, overtime pay, discretionary annual or performance bonuses, and sales incentives.
- Benefits may include medical, dental, and vision insurance.
- Paid time off is provided.
Categories
ML EngineeringRobotics
About Zipline
Zipline designs, builds, and operates autonomous drone delivery systems for healthcare, retail, and public-sector customers. The privately held company runs end-to-end logistics—aircraft, software, and fulfillment—to deliver blood, vaccines, medicines, and consumer goods. Founded in 2014 and headquartered in South San Francisco, Zipline operates across multiple continents; notable work includes Rwanda’s national blood deliveries and U.S. home delivery pilots with Walmart.