Hybrid opportunity at
ZooxMachine Learning Engineer - 3D Sensor Simulation
Zoox is building a fully self-driving vehicle fleet and the simulation tools needed to test it. This full-time Machine Learning Engineer role sits within the 3D…
Career Tools
About This Role
Zoox is building a fully self-driving vehicle fleet and the simulation tools needed to test it. This full-time Machine Learning Engineer role sits within the 3D Sensor Simulation team, which builds in-the-loop simulation and synthetic training data for autonomous vehicle software. The work centres on making simulated lidar, radar and camera output more realistic, combining machine learning with 3D simulation from learned sensor models through to neural 3D reconstruction. The…
Job Description
Zoox is building a fully self-driving vehicle fleet and the simulation tools needed to test it. This full-time Machine Learning Engineer role sits within the 3D Sensor Simulation team, which builds in-the-loop simulation and synthetic training data for autonomous vehicle software. The work centres on making simulated lidar, radar and camera output more realistic, combining machine learning with 3D simulation from learned sensor models through to neural 3D reconstruction.
The position suits engineers who already have industry experience applying machine learning to sensor data and who are comfortable building and tuning training pipelines using Python, PyTorch or TensorFlow. The listing marks the role as hybrid. Experience with generative models or 3D reconstruction is described as a bonus, and so is earlier work on autonomy in vehicles, robotics or computer vision.
Day to day, the engineer would coordinate with synthetic data, hardware-in-the-loop simulation, simulation validation and perception colleagues, define measures of sensor realism, and mentor engineers and interns. The posting does not state a base salary, a specific office location or a list of benefits, so those details should be confirmed directly with the employer before applying.
Responsibilities
- Design, train and validate machine learning models that make lidar, radar and camera simulation output more lifelike
- Work with synthetic data and hardware-in-the-loop teams to combine 3D Gaussian splatting and NeRF reconstruction with sensor simulation
- Assemble training sets from both real recordings and simulated sensor output, including generative data augmentation
- Define sensor realism metrics with simulation validation and perception engineers and address the largest sim-to-real gaps
- Mentor engineers and interns and set standards for training and deploying models on sensor data
Requirements
- Three or more years building machine learning models with Python and either PyTorch or TensorFlow
- Industry experience applying machine learning to lidar, radar and/or camera data
- Solid grasp of customizing, tuning and validating training pipelines
Qualifications
- Experience with generative methods such as diffusion models, flow matching or VAEs
- Experience with 3D reconstruction (3D Gaussian splatting, NeRFs) or differentiable rendering
- Prior work on autonomy in vehicles, robotics or computer vision
Core Skills
Frequently Asked Questions
Answers are based only on the employer’s listing; where it doesn’t say, neither do we.
Is this role hybrid or fully on site?
The listing marks the role as hybrid. The posting does not name a specific office location.
What experience does the role require?
At least three years building ML models with Python and either PyTorch or TensorFlow, plus industry experience applying ML to lidar, radar and/or camera data.
Is a salary or pay range stated?
No. The posting does not mention a salary or pay range.
How should I apply?
The posting does not describe the application steps. It only mentions an assigned recruiter for accommodation requests.
Sample Interview Questions
AI-generated questions tailored to this specific role — a preview of the full practice set.