Zipline
Autonomy Droid Perception SWE - Onboard Systems
$200K–$280K
South San Francisco, California, USA +1 · Full-time · Posted Jul 21, 2026
- Location
- South San Francisco, California, USA
South San Francisco, California, United States - Experience
- 5+ years for Senior or 8+ years for Staff; Senior requires full ML lifecycle experience, while Staff requires experience influencing design across systems.
About the job
Build and deploy real-time 3D perception systems for Zipline’s delivery droids, from onboard model optimization through evaluation tools and integration with motion planning.
What you'll do
- Implement, train, evaluate, and optimize multi-camera perception models for onboard computers.
- Build visualization, introspection, and evaluation tools to measure performance on datasets and in real-world deliveries.
- Coordinate with motion planning on subsystem interfaces, metrics, and production deployment.
What they ask for
- 5+ years (Senior) or 8+ years (Staff) building and deploying deep learning-based perception systems, particularly in 3D geometry, semantic segmentation, or learned multi-view stereo.
- Strong understanding of classical computer vision, including camera calibration, epipolar geometry, structure-from-motion, or 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.
Nice to have
- ML experience on hardware applications.
Pay
$200,000 - $240,000 USD base salary for Senior; $240,000 - $280,000 USD base salary for Staff
Tools
TensorRT, CNN, transformer, NVIDIA Jetson
AI focus
Develop and deploy real-time, multi-camera 3D perception ML models for autonomous delivery droids, optimizing models for resource-constrained onboard hardware.
Model-written summary · the original posting is authoritative