Autonomy Droid Perception SWE - Offboard Systems

What you'll do

• Train and deploy models for semantic segmentation, 4D geometry, and learned preference modeling using aerial, delivery, and synthetic data.
• Develop tools and high-fidelity priors to predict deliverability and ease customer onboarding in new environments.
• Build evaluation and validation infrastructure to verify model reliability in the field.

What they ask for

• At least 5+ years (Senior) or 8+ years (Staff) building and deploying deep learning-based perception systems, particularly in 3D geometry, semantic understanding, or mapping from remote sensing data.
• Strong understanding of classical computer vision and ability to combine it with modern ML approaches.
• Hands-on experience training, fine-tuning, iterating on, and optimizing CNN, transformer, and foundation-style architectures in production.
• Familiarity with training, data annotation, and evaluation pipelines; Staff candidates must have experience in real-world, high-stakes robotics or autonomy applications.

Nice to have

• Real-world, high-stakes robotics or autonomy application experience for Senior candidates.
• At Staff level, experience starting new efforts and building staged roadmaps for challenging ML tasks.

Pay

$200,000 - $240,000 USD base salary (Senior); $240,000 - $280,000 USD base salary (Staff)

Tools

CNN, transformer, foundation-style model architectures, feedforward multi-view models, gaussian splats

AI focus

Production perception ML for autonomous delivery, including semantic segmentation, 4D geometry, learned preference modeling, and model evaluation.