Machine Learning Engineer - 3D Sensor Simulation
At a glance
Mid-level Machine Learning Engineering role at Zoox. Foster City · full-time · $189,000–$283,000 base.
aimlroles summary, based on the employer's posting.
What you'll do
- Train and assess models that make lidar, radar and camera outputs more lifelike in autonomous-vehicle simulations
- Connect 3DGS and NeRF reconstruction methods to simulated sensor outputs alongside Synthetic Data and HIL Sim teams
- Prepare learning data from real-world and simulated sensors, expanding it with generative augmentation
- Set realism measures with simulation verification and perception specialists, then focus on the sim-to-real gaps that matter most
- Coach engineers and interns while setting sound methods for developing and deploying models on Zoox sensor data
What you bring
- At least three years developing machine learning models in Python using PyTorch or TensorFlow
- Professional experience applying machine learning to sensor inputs such as lidar, radar or cameras
- A firm understanding of how to customize, tune and validate model-training pipelines
Who this fits
This mid-level hybrid position in Foster City suits an engineer with sensor-data machine learning experience who wants to work across ML and 3D simulation. The role includes mentoring engineers and interns.
From the employer
The Zoox 3D Sensor Simulation team’s work is at the core of creating in-the-loop simulation and high fidelity synthetic data for training and testing autonomous vehicle AI. We are looking for a machine learning professional with experience applying ML to sensor data, such as LiDAR, radar, and cameras.
This role lies at the intersection of machine learning and 3D simulation at Zoox, ranging from learned sensor models to neural 3D reconstruction. By improving the realism of simulated sensors (lidar, radar, and cameras), you’ll enable the Zoox robotics autonomy stack to learn, improve, and enhance the mobility of humans.
In this role, you will:
-
Design, train, and validate ML models that improve the realism of simulated lidar, radar, and camera data in autonomous vehicle (AV) simulation pipelines.
-
Collaborate with Synthetic Data and HIL Sim teams to integrate 3DGS and NeRF-based 3D reconstruction with simulation of all sensor modalities.
-
Build and curate training datasets from real and simulated sensor data, including GenAI-based data augmentation.
-
Partner with Sim V&V and Perception engineers to define sensor realism metrics and close the sim-to-real gaps with the greatest impact on autonomy.
-
Mentor engineers and interns, and establish best practices for training and deploying ML on Zoox sensor data.
Qualifications:
- 3+ years of experience developing ML models in Python with PyTorch or TensorFlow
- Industry experience applying ML to sensor data (lidar, radar, and/or cameras)
- Solid understanding of customizing, tuning, and validating ML training pipelines
Bonus Qualifications:
- Experience with generative techniques such as diffusion, flow matching, or VAEs
- Experience with 3D reconstruction (3DGS, NeRFs) or differentiable rendering
- Prior experience developing autonomy in AVs or other robotics and computer vision
Follow us on LinkedIn
AccommodationsIf you need an accommodation to participate in the application or interview process please reach out to accommodations@zoox.com or your assigned recruiter.
A Final Note:You do not need to match every listed expectation to apply for this position. Here at Zoox, we know that diverse perspectives foster the innovation we need to be successful, and we are committed to building a team that encompasses a variety of backgrounds, experiences, and skills.