Motion Planning Engineer, TeleGuidance

12 Minutes ago • All levels • $214,000 PA - $270,000 PA

Research Development

Job Description

The TeleGuidance Planning team at Zoox enables autonomous vehicles to incorporate human contextual guidance into planning algorithms, allowing them to navigate complex scenarios by leveraging real-time remote human guidance. This role involves developing planner behavior and trajectories in collaboration with remote guidance, writing on-vehicle C++ components, and solving challenging edge cases for autonomous systems crucial to Zoox’s mission.

Good To Have:

PhD in Computer Science or related field
Experience with autonomous robots, Telerobotics
Experience with CUDA, Protobuf, or TypeScript

Must Have:

Master's degree in Computer Science or related field
Experience writing performance-critical production code in C++
Demonstrated ability to create real-time motion planning algorithms
Understanding of configuration spaces and a variety of planning techniques (A*, D*, RRTs, etc.)
Extensive experience with programming and algorithm design
Strong mathematical and geometrical skills
Experience leading small projects or teams

Perks:

Paid time off (e.g. sick leave, vacation, bereavement)
Unpaid time off
Zoox Stock Appreciation Rights
Amazon RSUs
Health insurance
Long-term care insurance
Long-term and short-term disability insurance
Life insurance

Add these skills to join the top 1% applicants for this job

cpp

game-texts

quality-control

test-coverage

cuda

mathematical

algorithms

typescript

The TeleGuidance Planning team at Zoox enables our autonomous vehicles to incorporate human contextual guidance into the planning algorithms, allowing our vehicles to successfully reason about and navigate complex scenarios by leveraging real-time remote human guidance. TeleGuidance is an integral part of the self-driving stack, dealing with some of the most challenging scenarios by fusing the “human sensor” data in real-time to plan a solution to complete the mission.

Take a look at this video we made for a more thorough explanation: https://youtu.be/NKQHuutVx78

Zoox is looking for a motion planning software engineer to join our Planning and Control team. In this role, you will help us develop planner behavior and trajectories in collaboration with remote guidance and write on-vehicle C++ components that integrate human guidance into the execution of our robot. You’ll solve the most complex and challenging edge cases for autonomous systems that are absolutely crucial to enabling Zoox’s mission.

In this role, you will:

Write on-vehicle C++ code that seamlessly integrates guidance provided by remote operators into the autonomous planner stack.
Work with fellow Planner engineers from vertical teams (agent handling, junctions, etc.), and you will represent TeleGuidance as a member on horizontal teams (construction zones, emergency vehicles, etc.).
Design and implement new tools and features that remote operators will use to help our fleet of vehicles to navigate complex dense urban driving environments. You will work with QA to develop a test plan for your new features.
Have the opportunity to contribute to a broad portion of the code base, as dictated by your skills and interests. Members of our team have contributed to motion planning, trajectory optimization, GPU code, offline tooling/simulation, cost tuning, and more.
Collaborate with developers across the AI software organization as a whole: Prediction, Perception, Core/Infra, and more.

Qualifications

Master's degree in Computer Science or related field
Experience writing performance-critical production code in C++
Demonstrated ability to create real-time motion planning algorithms
Understanding of configuration spaces and a variety of planning techniques (A*, D*, RRTs, etc.)
Extensive experience with programming and algorithm design
Strong mathematical and geometrical skills
Experience leading small projects or teams